.
The LIHEAP Formula
Libby Perl
Specialist in Housing Policy
June 26, 2015
Congressional Research Service
7-5700
www.crs.gov
RL33275
c11173008
The LIHEAP Formula
.
Summary
The Low Income Home Energy Assistance Program (LIHEAP) provides funds to states, the
District of Columbia, U.S. territories and commonwealths, and Indian tribal organizations
(collectively referred to as grantees) primarily to help low-income households pay home energy
expenses. The LIHEAP statute provides for two types of funding: regular funds (sometimes
referred to as block grant funds) and emergency contingency funds. Regular funds are allocated to
grantees based on a formula, while emergency contingency funds may be released to one or more
grantees at the discretion of the Secretary of the Department of Health and Human Services based
on emergency need. This report focuses on the way in which regular funds are distributed.
Regular LIHEAP funds are allocated to the states according to a formula that has a long and
complicated history. (Tribes and territories receive funds through set asides.) In 1980, Congress
created the predecessor program to LIHEAP, the Low Income Energy Assistance Program
The LIHEAP Formula
February 23, 2016
(RL33275)
Jump to Main Text of Report
Summary
The Low Income Home Energy Assistance Program (LIHEAP) provides funds to states, the District of Columbia, U.S. territories and commonwealths, and Indian tribal organizations (collectively referred to as grantees) primarily to help low-income households pay home energy expenses. The LIHEAP statute provides for two types of funding: regular funds (sometimes referred to as block grant funds) and emergency contingency funds. Regular funds are allocated to grantees based on a formula, while emergency contingency funds may be released to one or more grantees at the discretion of the Secretary of the Department of Health and Human Services based on emergency need. This report focuses on the way in which regular funds are distributed.
Regular LIHEAP funds are allocated to the states according to a formula that has a long and complicated history. (Tribes and territories receive funds through set asides.) In 1980, Congress created the predecessor program to LIHEAP, the Low Income Energy Assistance Program (LIEAP), as part of the Crude Oil Windfall Profits Tax Act (P.L. 96-223). Because Congress was
particularly concerned with the high costs of heating, funds under LIEAP were distributed
according to a multi-step formula that benefitted cold-weather states. In 1981, Congress enacted
LIHEAP as part of the Omnibus Budget Reconciliation Act (P.L. 97-35), replacing LIEAP.
However, the LIHEAP statute specified that states would continue to receive the same percentage
of regular funds that they did under the LIEAP formula (this is sometimes referred to as the
“old”
"old" LIHEAP formula).
When Congress reauthorized LIHEAP in 1984 as part of the Human Services Reauthorization Act
( (P.L. 98-558), it changed the program
’'s formula by requiring the use of more recent population
and energy data and requiring that HHS consider both heating and cooling costs of low-income
households (a change from what had largely been a focus on the need for heating assistance). The
effect of these changes meant that, in general, some funding would be shifted from cold-weather
states to warm-weather states. To prevent a dramatic shift of funds, Congress added two
“holdharmless”"hold-harmless" provisions to the formula. The percentage of funds that states receive under the
formula enacted in 1984 is sometimes referred to as the
“new” formula.
"new" formula.
The result of these provisions is a current law, three-tiered formula, the application of which
depends on the amount of regular funds that Congress appropriates. When appropriations are at or
below the equivalent of a hypothetical FY1984 appropriation of $1.975 billion, states receive the
“old” "old" formula percentage of funds. If appropriations exceed this level, then funds are allocated
according to the
“new”"new" formula percentage of funds, with certain states held harmless at the
level
level of funds they would have received at an appropriation of $1.975 billion in FY1984. Finally, when
appropriations are at or above $2.25 billion, there is a second hold-harmless provision in place, a
hold-harmless
raterate that ensures that certain states receive a set percentage of funds.
For many years after the enactment of the
“new”"new" LIHEAP formula, appropriations did not exceed
the equivalent of an FY1984 appropriation of $1.975 billion, so funds were distributed according
to the
“old”"old" formula percentages. However, in FY2006, and in FY2009 through
FY2015FY2016, regular
fund appropriations have ranged from $2.5 billion to $4.5 billion, and the
“new”"new" formula has been
incorporated into the way in which funds are distributed to the states.
In FY2016, the President’s
budget proposal and the House and Senate Appropriations Committee-passed bills would
incorporate the “new” LIHEAP formula into the fund distribution. For estimatedFor allocations to
the states, see Table C-1.
c11173008
Congressional Research Service
The LIHEAP Formula
.
Contents
Introduction...................................................................................................................................... 1
Predecessor Programs to LIHEAP ................................................................................................... 2
Community Services Administration Energy Assistance Programs .......................................... 2
The Low Income Energy Assistance Program (LIEAP) Formula ............................................. 5
Enactment of LIHEAP ..................................................................................................................... 9
Continued Use of the LIEAP Formula ...................................................................................... 9
The 1984 LIHEAP Reauthorization: A New Formula ............................................................... 9
Formula Discussions ........................................................................................................... 9
Introduction of a Hold-Harmless Level............................................................................. 10
Introduction of a Hold-Harmless Rate .............................................................................. 11
LIHEAP Formula Statutory Language .............................................................................. 11
Determining LIHEAP Regular Fund Allotments Using the “New” Formula ................................ 12
Calculating the New Formula Rates ........................................................................................ 13
Using the New Formula Percentages to Allocate Funds to the States ..................................... 15
“Old” Formula: Appropriations At or Below $1.975 Billion ............................................ 15
“New” Formula with Hold-Harmless Level: Appropriations Between $1.975
Billion and $2.25 Billion ................................................................................................ 16
“New” Formula with Hold-Harmless Level and Rate: Appropriations At or Above
$2.25 Billion................................................................................................................... 17
Implementation of the “New” LIHEAP Formula .................................................................... 18
Figures
Figure B-1. Estimated LIHEAP Allocations at Various Hypothetical Appropriations
Levels for Three Types of States ................................................................................................ 28
Tables
Table 1. Factors Used in Select Energy Assistance Formulas, FY1975-FY1980 ............................ 4
Table 2. Distribution of Funds Under LIEAP .................................................................................. 8
Table 3. Low-Income Home Energy Program (LIHEAP): “Old” and “New” Allotment
Percentages by State, FY2015 .................................................................................................... 18
Table 4. Recent State Allotment Percentages Under the “New” LIHEAP Formula ...................... 20
Table A-1. LIHEAP Estimated State Allotments for Regular Funds at Various
Hypothetical Appropriation Levels............................................................................................. 24
Table C-1. LIHEAP Actual State Regular Fund Allocations for FY2009 through FY2015
and Estimated Allocations for FY2016....................................................................................... 31
c11173008
Congressional Research Service
The LIHEAP Formula
.
Appendixes
Appendix A. Estimated Allotments to the States Under Various Hypothetical
Appropriations Levels ................................................................................................................ 23
Appendix B. Further Depiction of How State Allotments Depend Upon Appropriation
Levels.......................................................................................................................................... 27
Appendix C. LIHEAP Regular Fund Allocations to the States, FY2009-FY2015, and
Estimated FY2016 Allocations ................................................................................................... 29
Contacts
Author Contact Information........................................................................................................... 35
c11173008
Congressional Research Service
The LIHEAP Formula
.
Introduction
the states, see Table C-1.
The LIHEAP Formula
Introduction
The Low Income Home Energy Assistance Program (LIHEAP) is a block grant
program
program administered by the Department of Health and Human Services (HHS) under which the federal
government gives annual grants to states, the District of Columbia, U.S. territories and
commonwealths, and Indian tribal organizations to operate multi-component home energy
assistance programs for needy households.
11 Established in 1981 by Title XXVI of P.L. 97-35, the
Omnibus Budget Reconciliation Act, LIHEAP has been reauthorized and amended a number of
times, most recently in 2005, when P.L. 109-58, the Energy Policy Act, authorized annual regular
LIHEAP funds at $5.1 billion per year from FY2005 through FY2007.
2
2
The federal LIHEAP statute has very broad guidelines, with many decisions regarding the
program’ program's operation made by the states. Recipients may be helped with their heating and cooling
costs, receive crisis assistance, have weatherizing expenses paid, or receive other aid designed to
reduce their home energy needs. Households with incomes up to 150% of the federal poverty
income guidelines or, if greater, 60% of the state median income, are federally eligible for
LIHEAP benefits. States may adopt lower income limits, but no household with income below
110% of the poverty guidelines may be considered ineligible. The most recent HHS data show
that an estimated 8.1 million households received winter heating or winter crisis assistance in
FY2010 (the largest share of LIHEAP funds pay for heating assistance).
3
3
The LIHEAP statute provides for two types of program funding: regular funds—sometimes
referred to as block grant funds—and emergency contingency funds. Regular funds are allotted to
states on the basis of the LIHEAP statutory formula, which was enacted as part of the Human
Services Reauthorization Act of 1984 (P.L. 98-558
).4).4 The way in which regular funds are
allocated to states depends on the amount of funds appropriated by Congress. The second type of
LIHEAP funds, emergency contingency funds, may be released and allotted to one or more states
at the discretion of the President and the Secretary of HHS.
55 The funds may be released at any
point in the fiscal year to meet additional home energy assistance needs created by a natural
disaster or other emergency.
6
6
The remainder of this report discusses only the history and methods of distributing regular
LIHEAP funds to the state. Funds for tribes are included in each state
’'s formula allocations and
are distributed at the state level based on eligible tribal members. Territories receive funds
separately as a percentage set aside of regular funds, so neither tribes nor territories are included
in the formula discussion.
1
For additional information on LIHEAP, see CRS Report RL31865, LIHEAP: Program and Funding, by Libby Perl.
LIHEAP is codified at 42 U.S.C. §§8621-8630.
3
U.S. Department of Health and Human Services, Administration for Children and Families, FY2010 LIHEAP Report
to Congress, p. 35.
4
The formula section is codified at 42 U.S.C. §8623.
5
Depending on how Congress appropriates them, contingency funds may remain available for distribution in more than
one fiscal year or they may expire with the fiscal year for which they were appropriated.
6
The statutory definition of emergency includes a significant home energy supply shortage or disruption, a significant
increase in the cost of home energy, a significant increase in home energy disconnections, a significant increase in
participation in a public benefit program, a significant increase in unemployment, or an event meeting such criteria as
the Secretary determines to be appropriate. 42 U.S.C. §8622.
2
c11173008
Congressional Research Service
1
The LIHEAP Formula
.
in the formula discussion.
Predecessor Programs to LIHEAP
The mid- to late-1970s, a time marked by rapidly rising fuel prices, also marked the beginning of
federal energy assistance funding for low-income households. The first national program to help
low-income households was created in early 1975 to assist families with energy conservation
primarily through home weatherization. This assistance was provided through a new Emergency
Energy Conservation Program (EECP), enacted as part of the Headstart, Economic Opportunity,
and Community Partnership Act of 1974 (P.L. 93-644). The funds were administered by the
Community Services Administration (CSA), the successor agency to the Office of Economic
Opportunity, which was responsible for many of the programs created as part of the 1964 war on
poverty. Beginning in 1977, funds were also made available through the CSA to help families
directly pay for fuel (as opposed to weatherization expenses) via a variety of programs. Each of
these programs had in common a focus on the need for heating assistance (versus cooling
assistance).
Congress continued to appropriate funds for energy assistance programs through FY1980, at
which point a new program, the Low Income Energy Assistance Program (LIEAP), was enacted
as part of the Crude Oil Windfall Profits Tax Act of 1980 (P.L. 96-223). LIEAP, which was
administered by the Department of Health and Human Services (HHS), was funded for one year,
FY1981, before the creation of LIHEAP. Like the CSA programs, LIEAP emphasized heating
over cooling needs. This preference was reflected in both the CSA program formulas and the
LIEAP set of formulas, which used variables that benefitted cold-weather states to determine how
funds would be distributed. The LIEAP set of formulas continues to have relevance for the way in
which LIHEAP funds are distributed. This section of the report describes these predecessor
programs to LIHEAP and their distribution formulas.
Community Services Administration Energy Assistance Programs
On January 4, 1975, President Ford signed into law the Headstart, Economic Opportunity, and
Community Partnership Act of 1974 (P.L. 93-644), which contained funds for a new program,
called the Emergency Energy Conservation Program (EECP). The program was to be
administered by the Community Services Administration (CSA), and its purpose was
to enable low-income individuals and families, including the elderly and the near poor, to
participate in energy conservation programs designed to lessen the impact of the high cost of
energy ... and to reduce ... energy consumption.
The law governing EECP listed a number of eligible activities in which states could participate,
including energy conservation and education programs; weatherization assistance; loans and
grants for the purchase of energy conservation technologies; alternative fuel supplies; and fuel
voucher and stamp programs. Despite the variety of activities that could be funded through the
program, the first CSA funding notice regarding the program limited eligible activities to
“winterizing” "winterizing" homes and to giving emergency assistance
“"to prevent hardship or danger to health
c11173008
Congressional Research Service
2
The LIHEAP Formula
.
due to utility shutoff or lack of fuel.
”7"7 During the four years the EECP was funded, the majority
of funds were used for weatherization expenses.
8
8
EECP funds were distributed to states via a formula that benefitted those states with high heating
costs. One formula variable in particular, a measure of
“coldness”"coldness" called heating degree days,
benefitted cold-weather states. Heating degree days measure the extent to which a day
’'s average
temperature falls below 65° Fahrenheit. For example, a day with an average temperature of 50°
results in a measure of 15 heating degree days. Because heating degree days are higher in
coldweathercold-weather states, including the heating degree day variable in a formula favors states with greater
heating needs. Squaring the heating degree days magnifies this effect.
99 The EECP formula took
the number of population-weighted heating degree days in each state, squared them, and
multiplied the result by the number of households in poverty that owned their homes to determine
how funds would be allocated.
1010 The CSA acknowledged the emphasis on heating needs in its
formula, stating that the FY1975 allocation
“"was heavily weighted to the coldest areas.
”11"11 In the
three fiscal years that followed the first appropriation for the EECP, from FY1976 through
FY1978, the CSA changed somewhat the way in which it allocated funds to the states; however,
the factors continued to favor cold-weather states through use of either heating degree days or
heating degree days squared.
12
12
The first year that Congress specifically appropriated funds for direct assistance to help
lowincomelow-income households (those at or below 125% of poverty) pay their energy costs (instead of funds
that went primarily for weatherization and conservation activities) was FY1977. The FY1977
Supplemental Appropriations Act (P.L. 95-26) provided $200 million for a Special Crisis
Intervention Program to be administered by CSA. States could use funds to make direct payments
to fuel providers on behalf of low-income families lacking the financial resources to pay their
energy bills. The CSA directed states to target households where utilities had been shut off (or
were threatened with shut off) or who could prove
“"dire financial need
”" as the result of paying
large energy bills.
1313 Although the law did not reserve funds exclusively for heating costs, the way
in which funds were allocated to the states emphasized heating need. Funds were distributed to
the states based on a formula that used (1) heating degree days squared, (2) the number of
households in poverty, (3) the number of persons above age 65 with incomes below 125% of
poverty, and (4) the relative cost of fuel in the region.
14 Congress again appropriated $200 million
7
Community Services Administration, “Character and Scope of Specific Community Action Programs: Emergency
Energy Conservation Program,” Federal Register, vol. 40, no. 145, July 28, 1975, p. 31603.
8
See, for example, House Appropriations Committee, report to accompany H.R. 4877, the FY1977 Supplemental
Appropriations Act, 95th Cong., 1st sess., H.Rept. 95-68, March 11, 1977: “The funds in this program are used primarily
to purchase materials to insulate the homes of low-income families.”
9
For example, if a southern state experiences 700 heating degree days in a year and a northern state experiences 7,000,
the northern state has 10 times as many heating degree days as the southern state. However, if both numbers are
squared, the northern state has 100 times as many heating degree days as the southern state.
10
Community Services Administration, “Emergency Energy Conservation Program: Submission of Funding Plans,”
Federal Register, vol. 41, no. 208, October 27, 1976, p. 47096.
11
Ibid.
12
Ibid., pp. 47096-47097.
13
Community Services Administration, “Special Crisis Intervention Program: General Information, Application
Procedures, and Post Grant Requirements,” Federal Register, vol. 42, no. 125, June 29, 1977, p. 33240.
14
The formula was described in the Senate Appropriations Committee report to accompany H.R. 4877, the FY1977
Supplemental Appropriations Act, 95th Cong., 1st sess., S.Rept. 95-64, March 24, 1977. The CSA implemented this
formula, which it described in guidance to the states. See the Federal Register, Ibid.
c11173008
Congressional Research Service
3
The LIHEAP Formula
.
14 Congress again appropriated $200 million for crisis intervention in both FY1978 and FY1979.
1515 In FY1978, funds were available to
households with the need for assistance as the result of an energy-related emergency such as lack
of fuel, a natural disaster, fuel shortages, and widespread unemployment.
1616 In FY1979, funds
were made available to assist families facing
“"substantially increased energy costs and/or life- or
health-threatening situations caused by winter-related energy emergencies.
”17
"17
In FY1980, Congress appropriated a total of $1.6 billion for energy assistance. Of this amount,
$400 million was appropriated for the Energy Crisis Assistance Program (ECAP, a CSA program
similar to the Special Crisis Intervention Program) through two separate appropriations.
1818 The
remainder, $1.2 billion, was appropriated as part of the FY1980 Department of the Interior
Appropriations Act (P.L. 96-126) to the Department of Health, Education, and Welfare (HEW, the
predecessor to HHS) for cash assistance and crisis intervention due to high energy costs. This
appropriation to HEW is sometimes referred to as Low Income Supplemental Energy Allowances.
Of this $1.2 billion, $400 million was to be distributed specifically to recipients of Supplemental
Security Income (SSI). The rest of the funds appropriated to HEW, approximately $800 million,
as well as the ECAP funds, were distributed to states on the basis of three factors: heating degree
days squared, the number of households below 125% of poverty, and the difference in home
heating energy expenditures between 1978 and 1979. The formula used to distribute the $400
million for SSI recipients used these same factors but also included the number of SSI recipients
in each state relative to the national total.
Table 1. Factors Used in Select Energy Assistance Formulas, FY1975-FY1980
Emergency Energy
Conservation Program:
a
FY1975
(P.L. 93-644)
Special Crisis
Intervention Program:b
FY1977
(P.L. 95-26)
Low Income Supplemental
Energy Allowances:c
FY1980
(P.L. 96-126)
(Heating degree days)2
(Heating degree days)2
(Heating degree days)2
Number of homeowners in
poverty
Number of households in poverty
Number of households below 125% of
poverty
Number of persons over age 65 with
income less than 125% of poverty
Difference in home heating
expenditures between 1978 and 1979
Relative cost of fuel
Sources:aFY1975 (P.L. 93-644)
Special Crisis Intervention Program:bFY1977 (P.L. 95-26)
Low Income Supplemental Energy Allowances:cFY1980 (P.L. 96-126)
(Heating degree days)2
|
(Heating degree days)2
|
(Heating degree days)2
|
Number of homeowners in poverty
|
Number of households in poverty
|
Number of households below 125% of poverty
|
Number of persons over age 65 with income less than 125% of poverty
|
Difference in home heating expenditures between 1978 and 1979
|
Relative cost of fuel
|
Sources: For the formula under P.L. 93-644, see Community Services Administration,
“"Emergency Energy
Conservation Program: Submission of Funding Plans,
” " Federal Register, vol. 41, no. 208, October 27, 1976,
p. p. 47096. For the formula under P.L. 95-26, see Senate Appropriations Committee, report to accompany H.R.
4877, the FY1977 Supplemental Appropriations Act,
95th Congress, 1st95th Congress, 1st session, S.Rept. 95-64, March 24, 1977.
The formula for P.L. 96-126
is contained within the law.
a.
Of the funds appropriated for the Emergency Energy Conservation Program, 90% were distributed via the formula, while the remaining 10% were divided among the 12 coldest states as measured by heating degree days. The formula involved multiplying heating degree days squared by the number of homeowners in poverty to arrive at the percentage share for each state.
b.
The Special Crisis Intervention Program did not specify a weight for each of the four variables used to determine allocations.
c.
The Low Income Supplemental Energy Allowances arrived at states' shares of funds through the formula ½ (heating degree days2 * number of households below 125% of poverty) + ½ (difference in home heating expenditures between 1978 and 1980). Of the $1.6 billion appropriated for energy assistance in FY1980, $400 million was set aside for SSI recipients. The formula to distribute those funds was ⅓ (heating degree days2 * number of households below 125% of poverty) + ⅓ (difference in home heating expenditures between 1978 and 1979) + ⅓ (SSI recipients in each state relative to the national total).
The Low Income Energy Assistance Program (LIEAP) Formula
In April 1980, Congress replaced the patchwork energy assistance programs of the late 1970s with one program, the Low Income Energy Assistance Program (LIEAP). LIEAP, the direct predecessor program to LIHEAP, was established as part of the Crude Oil Windfall Profits Tax Act of 1980 (P.L. 96-223). The program was introduced in the Senate as the Home Energy is contained within the law.
15
Funds were appropriated through the FY1978 Supplemental Appropriations Act (P.L. 95-240) and in FY1979
through a continuing resolution (P.L. 95-482). In FY1978, Congress called the program Emergency Energy Assistance
Program and in FY1979 called it the Crisis Intervention Program (excluding the word “Special” from the title).
16
Community Services Administration, “Emergency Energy Conservation Program: Funding Requirements for
Emergency Energy Assistance Program,” Federal Register, vol. 43, no. 46, March 8, 1978, p. 9476.
17
Community Services Administration, “Emergency Energy Conservation Program: Fiscal Year 1979 Crisis
Intervention Program,” Federal Register, vol. 43, no. 250, December 28, 1978, pp. 60466-60467.
18
Congress appropriated $250 million for ECAP as part of an FY1980 Continuing Resolution (P.L. 96-123, referencing
the FY1980 Departments of Labor, Health and Human Services and Education Appropriations bill, H.R. 4389), and
appropriated an additional $150 million as part of the Department of the Interior Appropriations Act (P.L. 96-126).
c11173008
Congressional Research Service
4
The LIHEAP Formula
.
a.
Of the funds appropriated for the Emergency Energy Conservation Program, 90% were distributed via the
formula, while the remaining 10% were divided among the 12 coldest states as measured by heating degree
days. The formula involved multiplying heating degree days squared by the number of homeowners in
poverty to arrive at the percentage share for each state.
b.
The Special Crisis Intervention Program did not specify a weight for each of the four variables used to
determine allocations.
c.
The Low Income Supplemental Energy Allowances arrived at states’ shares of funds through the formula ½
(heating degree days2 * number of households below 125% of poverty) + ½ (difference in home heating
expenditures between 1978 and 1980). Of the $1.6 billion appropriated for energy assistance in FY1980,
$400 million was set aside for SSI recipients. The formula to distribute those funds was ⅓ (heating degree
days2 * number of households below 125% of poverty) + ⅓ (difference in home heating expenditures
between 1978 and 1979) + ⅓ (SSI recipients in each state relative to the national total).
The Low Income Energy Assistance Program (LIEAP) Formula
In April 1980, Congress replaced the patchwork energy assistance programs of the late 1970s
with one program, the Low Income Energy Assistance Program (LIEAP). LIEAP, the direct
predecessor program to LIHEAP, was established as part of the Crude Oil Windfall Profits Tax
Act of 1980 (P.L. 96-223). The program was introduced in the Senate as the Home Energy
Assistance Act (S. 1724) and was incorporated into H.R. 3919, the bill that would become the
Crude Oil Windfall Profits Tax Act, on the Senate floor.
1919 Like the energy assistance programs of
the late 1970s such as the Special Crisis Intervention Program and the Low Income Supplemental
Energy Allowances, LIEAP allocated funds to states in order to help low-income households pay
their home energy costs. Also like these predecessor programs, LIEAP allocated funds to states
using a method that put more emphasis on the heating needs of cold-weather states than it did on
cooling needs.
The formula developed under LIEAP continues to be relevant in several ways: (1) it has been
used to distribute LIHEAP funds as recently as FY2007, (2) the percentage shares of funds that
states received continue to be the benchmark for the way in which states are held harmless under
the current LIHEAP formula, and (3) from FY2009 through FY2012, Congress has distributed the
bulk of LIHEAP funds using the LIEAP formula percentages (for more information, see
Appendix C). As a result, the variables used are important in understanding the current formula
and the way in which it is used to distribute funds.
Ultimately, Congress developed the LIEAP formula through two different laws: P.L. 96-223, the
law that authorized LIEAP, and P.L. 96-369, a continuing resolution enacted six months later. The
following two subsections describe the elements of the formula developed through each.
Formula Under P.L. 96-223
The formula developed as part of S. 1724, and subsequently incorporated into P.L. 96-223
,
, reflected, in part, the concern that the problem of rising energy costs were
“"most critical in areas
with high home heating costs.
”20"20 The formula for LIEAP arose from a Senate compromise over
three different proposals. The debate centered around the degree to which heating should be
19
“Windfall Profits Tax.” In CQ Almanac 1979, 35th ed., 609-32 (Washington, DC: Congressional Quarterly, 1980)
http://library.cqpress.com/cqalmanac/cqal79-1184031.
20
Senate Committee on Labor and Human Resources, Home Energy Assistance Act, report to accompany S. 1724, 96th
Cong., 1st sess., S.Rept. 96-378, October 25, 1979, p. 12.
c11173008
Congressional Research Service
5
The LIHEAP Formula
.
emphasized over energy expenditures generally. Some Members wanted a formula that accounted
for all energy uses and was not based solely on geographic location,
2121 while others saw the
program’ program's purpose as solely to provide heating assistance.
2222 The debate on the Senate floor was,
at times, contentious, with Senator Edmund Muskie (ME) resolved to filibuster in order to
support the heating needs of northern states.
2323 Primarily at issue was the measure of heating
degree days, particularly the extent to which they would be weighted and whether they would be
squared.
Under the final compromise LIEAP formula in P.L. 96-223, states received funds under one of
four different alternatives used to measure home energy need, depending on which one benefitted
a state the most. Three of the four options contained different combinations of several formula
factors: residential energy expenditures; heating degree days or heating degree days squared; and
the number of low-income households in the state.
•
Under the first formula alternative, 50% of the allocation was based on
residential energy expenditures and 50% on heating degree days squared
multiplied by the number of households at or below the Bureau of Labor
Statistics (BLS) lower living standard.
24
•
24
Under the second formula alternative, 25% of the allocation was based on
residential energy expenditures and 75% based on heating degree days squared
multiplied by the number of households at or below the BLS lower living
standard.
•
Under the third formula alternative, 50% of the allocation was based on
residential energy expenditures and 50% based on heating degree days (not
squared) multiplied by the number of households with incomes at or below the
BLS lower living standard.
•
The fourth option guaranteed states a minimum benefit of $120 for each
household that received Aid to Families with Dependent Children (AFDC), SSI,
or Food Stamp benefits. The option was added to S. 1724
at the Finance Committee level in recognition of the fact that (in general) funds were not being provided for cooling costs.25 (See Table 2 for a breakdown of these formulas.)
While the focus of the formula was on heating assistance, the LIEAP law did allow states to provide for cooling when households could demonstrate medical necessity.26 Congress authorized at the Finance
21
See, for example, Senator Russell Long, “Home Energy Assistance Act,” Senate debate, Congressional Record, vol.
125, part 25 (November 14, 1979), p. 32278. “But the formula [as passed by the Senate Finance Committee] went a
long way toward considering the total household expense for energy, not just heating.”
22
Senator Rudy Boschwitz, “Home Energy Assistance Act,” Senate debate, Congressional Record, vol. 125, part 25
(November 14, 1979), p. 32290. “I refer back to the committee report, which talks about the intent of the act being to
‘offset high heating costs (and cooling where medically necessary) and that assistance not be a supplement of all
utilities and their use to run appliances, etc.’... It is very clear that it is the intent of the Senate to help keep people
warm.”
23
Senator Edmund Muskie, “Home Energy Assistance Act,” Senate debate, Congressional Record, vol. 125, part 25
(November 14, 1979), p. 32288. “I do not often do this. As a matter of fact, this is my 21st year in the Senate, and I can
recall only one other time in which I have sought to use delay and extended debate to make a point and to achieve
justice. I am not a filibusterer. If I did not believe deeply about this, I would not be standing here.”
24
The BLS determined the lower living standard income level through its annual family budgets, which it maintained
from 1947 to 1981. At the time the LIEAP program was enacted, the BLS developed annual family budgets assuming
three different standards of living: lower, intermediate, and higher. The budget was calculated using costs of consumer
goods including food, housing, transportation, clothing, and health care (unlike the federal poverty guidelines, which
are based on the amount of money needed to buy food). The budget was then adjusted for family size and the prices of
goods in various cities throughout the country. See David S. Johnson, John M. Rogers, and Lucilla Tan, “A Century of
Family Budgets in the United States,” Monthly Labor Review, 124, no. 5 (May 2001): 28-45.
c11173008
Congressional Research Service
6
The LIHEAP Formula
.
Committee level in recognition of the fact that (in general) funds were not being
provided for cooling costs.25
(See Table 2 for a breakdown of these formulas.)
While the focus of the formula was on heating assistance, the LIEAP law did allow states to
provide for cooling when households could demonstrate medical necessity.26 Congress authorized
LIEAP for one year, FY1981, at $3 billion, but funds were not appropriated as part of
P.L. 96-223.
P.L. 96223.
Formula Under P.L. 96-369
Before the formula in P.L. 96-223 could be used to allocate funds, Congress introduced an
alternative method for computing the state distribution rates. It did so when it appropriated $1.85
billion in LIEAP funds for FY1981 in a continuing resolution (P.L. 96-369), in October of 1980,
six months after enactment of the Crude Oil Windfall Profits Tax Act. The new allocation method
was not described in P.L. 96-369, however. Instead, the continuing resolution referred to a House
Appropriations Committee report (H. Rept. 96-1244) accompanying another bill—the FY1981
Departments of Labor, Health and Human Services and Education Appropriations Act. It was in
this committee report that the additional formula components for LIEAP were laid out.
2727 The
additional formula components appeared to be intended to act as a counter to the formula
developed in P.L. 96-223, which some argued benefitted warmer weather states more than was
necessary.
28
28
The first step in the new set of formulas was to determine each state
’'s share of funds using two
calculations set out in H. Rept. 96-1244 and assign states the greater of the two amounts.
•
Under the first formula alternative, 50% of the allocation was based on the
increase in home heating expenditures, and 50% was based on the number of
heating degree days squared times the population with income less than or equal
to 125% of poverty. This was the same formula used for the Low-Income
Supplemental Energy Allowances Program.
25
Senator Russell B. Long, “Home Energy Assistance Act,” Senate debate, Congressional Record, vol. 125, part 25
(November 15, 1979), p. 32561. “This language was evolved in the Finance Committee. When the majority of the
committee voted to exclude such items as air-conditioning and anything related to cooling a house and limited that
formula to heating, this Senator contended that, if that were to be the case, there should be at least a minimum on which
people could depend.”
26
According to the law, “The State is authorized to make grants to eligible households to meet the rising costs of
cooling whenever the household establishes that such cooling is the result of medical need pursuant to standards
established by the Secretary.”
27
House Committee on Appropriations, report to accompany H.R. 7998, the FY1981 Departments of Labor, Health
and Human Services, and Education Appropriations Act, 96th Cong., 2nd sess., H. Rept. 96-1244, August 21, 1980, pp.
75-76.
28
See, for example, Representative David Obey, “Low Income Energy Assistance,” House debate, Congressional
Record, vol. 126, part 18 (August 27, 1980), p. 23505. “Last year the Congress adopted a formula which, very frankly,
was unfair to the South. It provided a much larger amount of the money available than it should have to Northern
States. In response to that, Senator Long, on the windfall profit tax legislation, adopted an amendment which, for the
block grant portion of the program, provided phenomenal increases for the Southern States at the expense of the
Northern States.”
c11173008
Congressional Research Service
7
The LIHEAP Formula
.
•
Supplemental Energy Allowances Program.
Under the second formula alternative, 25% of the allocation was based on total
residential energy expenditures, and 75% was based on heating degree days
squared multiplied by the number of low-income households in the state.
The greater of the two percentages calculated using the formula in H. Rept. 96-1244 was then
assigned to each state. After adjusting state allotments proportionately so that the total allocation
reached 100% of funds available, the second step in the amended formula was to compare these
state allotments to 75% of the amount each state would receive under the formula in P.L. 96-223
.
. States would then receive the greater of these two amounts. To see the percentage of funds that
each state received under the LIEAP formula, see Table 3, column (a).
Although the alternative formulas under H.Rept. 96-1244 used factors similar to those in P.L.
9622396-223, the original set of formulas was somewhat more favorable to warm-weather states. For
example, the BLS lower living standard, used in all of the P.L. 96-223 formulas but only one of
those in H.Rept. 96-1244, was higher than 125% of poverty for most household sizes, which
benefitted the South, where the low-income population was higher.
2929 The original set of formulas
in P.L. 96-223 also provided for a minimum benefit to states on the basis of the number of AFDC,
SSI, and Food Stamp recipient households, unconditioned on their household heating
expenditures. In addition, the inclusion of the increase in home heating expenditures in H. Rept.
96-1244 benefitted Northeastern states, where heating oil prices had increased substantially.
30
30
Table 2. Distribution of Funds Under LIEAP
P.L. 96-223
P.L. 96-369
P.L. 96-223
|
P.L. 96-369
|
Assign each state the option under which they receive
the greatest proportion of funds. If Options 2 and 3
both result in a greater proportion than Option 1,
assign the state the lesser of Option 2 or
3.
3.
Each state receives the greater of 75% of the
amount under P.L. 96-223 or Option 1 or Option
2 under
P.L. 96-369.
Option 1:
|
½
|
Residential energy expenditures
|
Option 1:
|
½
|
Increase in home heating expenditures from 1978-1980a
½
|
P.L. 96-369.
Option 1:
Option 1:
½ Residential energy expenditures
½ (Heating degree days)
22 * Households with
income ≤ BLS lower living standard
Option 2:
¼ Residential energy expenditures
¾
½
|
(Heating degree days)
2 * Population with income ≤ 125% of poverty
Option 2:
|
¼
|
Residential energy expenditures
|
Option 2:
|
¼
|
Total residential energy expenditures 1980
|
¾
|
(Heating degree days)2 * Households with income ≤ BLS lower living standard
|
¾
|
2 * Households with
income ≤ BLS lower living standard
Option 3:
½ Increase in home heating expenditures
from 1978-1980a
½ (Heating degree days)2 * Population with
income ≤ 125% of poverty
Option 2:
¼ Total residential energy expenditures
1980
¾ (Heating degree days)
2 2 * Households
with income ≤ BLS lower living standard
½ Residential energy expenditures
½ Heating degree days * Households with
income ≤ BLS lower living standard
Option 4:
Funds sufficient for a minimum benefit of
$120 per AFDC, SSI, and Food Stamprecipient household
29
“The Low-Income Home Energy Assistance Program: An Analysis of the 1984 Reauthorization Issues,” Coalition of
Northeastern Governors, April 1984, p. 5.
30
H.Rept. 96-1244 did not specify the years between which the increase in home heating expenditures should be
measured. In implementing the formula, HHS measured the increase between 1978 and 1980.
c11173008
Congressional Research Service
8
The LIHEAP Formula
.
Source:
Option 3:
|
½
|
Residential energy expenditures
|
½
|
Heating degree days * Households with income ≤ BLS lower living standard
|
Option 4:
|
Funds sufficient for a minimum benefit of $120 per AFDC, SSI, and Food Stamp- recipient household
|
Source: The Crude Oil Windfall Profits Tax Act (P.L. 96-223) and the House Appropriations Committee
Report to Accompany H.R. 7998, the FY1981 Departments of Labor, Health and Human Services, and Education
Appropriations Bill, H.Rept. 96-1244, August 21, 1980.
Notes: * Multiplied by.
≤ Less than or equal to.
a.
a.
H.Rept. 96-1244 did not specify which years would be used to determine residential energy expenditures;
1978 and 1980 were the years used by HHS.
Enactment of LIHEAP
In August 1981, the Omnibus Budget Reconciliation Act, P.L. 97-35, created LIHEAP, replacing
its predecessor, LIEAP. The new program was not substantially different from the previous
program. Some of the changes to the program included less restrictive federal rules and more
state flexibility in determining how to operate their LIHEAP programs. The program was
authorized at $1.85 billion for FY1982-FY1984. In FY1982, Congress appropriated $1.875
billion for LIHEAP; in FY1983, it appropriated $1.975 billion; and in FY1984, $2.075 billion.
Continued Use of the LIEAP Formula
When the formula for LIEAP was initially created in 1980 under the Crude Oil Windfall Profits
Tax Act (P.L. 96-223), it brought about a good deal of debate on the floor of the Senate, where the
formula provisions were added to the legislation.
3131 Discussion over the formula also occurred
leading up to the enactment of P.L. 96-369, the FY1981 continuing resolution that funded LIEAP
and amended the formula.
3232 Despite these earlier disagreements over formula allocations, the
process to enact LIHEAP in 1981 did not engender the same level of debate or result in a different
formula. Instead, the law creating LIHEAP provided that the allotment percentages for each state
would remain the same as they had been in FY1981 under the LIEAP formula as amended by P.L.
96-369. From FY1982 through FY1984, then, states continued to receive the same percentage of
funds that they received under the LIEAP formula.
The 1984 LIHEAP Reauthorization: A New Formula
Formula Discussions
Formula Discussions
When Congress began to consider reauthorizing LIHEAP in 1983, two aspects of the formula
were debated. First, some legislators recognized that the multi-step LIEAP formula benefitted
cold-weather states relative to warm-weather states.
3333 The second debated aspect of the formula
centered on the appropriateness and timeliness of the data used in formula calculations. In 1983,
the energy information used to calculate state allotments was not the most current data
31
See, for example, Senate debate, Congressional Record, vol. 125, parts 24-25 (November 13-15, 1979), pp. 3208232086, 32275-32293, 32558-32565, and 32576-32589.
32
House debate, Congressional Record, vol. 126, part 18 (August 27, 1980), pp. 23502-23515.
33
See, for example, Comments of Rep. Billy Tauzin, U.S. Congress, Joint Hearing before the Subcommittees of the
Committees on Energy and Commerce, Education and Labor, and Ways and Means, Energy Costs and Low Income
Energy Assistance, 98th Cong., 1st sess., February 24, 1983, pp. 119-120.
c11173008
Congressional Research Service
9
The LIHEAP Formula
.
available.34 available.34 For example, the most recent data the formula used were the change in the cost of
energy between 1978 and 1980, or the cost of energy in 1980, depending on the sub-formula one
chose to apply. No aspect of the formula took account of increased costs after 1980.
35
35
Legislative sentiment in favor of changing the formula was evident, when, in September 1983,
the House adopted an amendment to the Emergency Immigration Education Act (H.R. 3520) that
would have adjusted the LIHEAP formula and resulted in a change in allocations to the states.
The amendment
’'s formula took into account the energy expenditures of poor families, which,
according to the amendment
’'s sponsor, Representative Carlos Moorhead (California), would
result in lower percentage allocations for 23 states, mostly in the Northeast and Midwest, gains
for 27, primarily in the South, and the same allocation for one state.
3636 The amendment was
eventually dropped from H.R. 3520 in conference with the Senate.
Introduction of a Hold-Harmless Level
Efforts to reauthorize LIHEAP began in April 1983 with the introduction of the Low-Income
Home Energy Assistance Amendments of 1984 (H.R. 2439). The bill was referred to two
committees: Education and Labor and Energy and Commerce. Within the Energy and Commerce
committee, two subcommittees held mark-ups: Fossil and Synthetic Fuels and Energy
Conservation and Power.
As introduced, H.R. 2439 did not contain changes to the LIHEAP formula. The Subcommittees
on Fossil and Synthetic Fuels and Energy Conservation and Power worked together to arrive at a
formula change, which had the effect of shifting funds from states in the Northeast to the South
and West. Unlike the previous set of formulas developed under LIEAP, the new formula directed
the Department of Health and Human Services to determine states
’' allotments
“"using data relating
to the most recent year for which data is available.
”" Because the cost of heating oil remained
steady between 1981 and 1983, and the price of natural gas rose 33%, this meant that states in the
Northeast—where heating oil was the primary source of energy—would lose LIHEAP dollars,
while states in the South and the Midwest would gain under this provision.
3737 In addition,
population growth in the South (as well as its higher poverty rates) meant that southern states
would benefit from the use of more recent population data.
To offset the losses to certain states resulting from the use of current data, H.R. 2439 also
included a hold-harmless provision, or hold-harmless level; this provision ensured that if
appropriations were less than or equal to $1.875 billion, states would receive no less than their
allotment would have been under the old formula at this appropriations level. The bill
additionally increased the LIHEAP authorization level to $2.075 billion for FY1984, $2.26 billion
for FY1985, $2.5 billion in FY1986, $2.625 billion for FY1987, and $2.8 billion for FY1988.
34
Report of the Committee on Energy and Commerce to accompany H.R. 2439, the Low-Income Home Energy
Assistance Amendments of 1984, 98th Cong., 2nd sess., H.Rept. 98-139, Part 2, May 15, 1984, p. 13.
35
Ibid., p. 4.
36
Congressional Record, vol. 129, part 17 (September 13, 1983), p. 23877. The greatest increases in percentage
allocations were for Florida at 51%, Texas at 44%, and Alabama at 37%. The states whose percentage allocations
decreased the most were Vermont at 32%, North Dakota at 24%, and New Hampshire at 23%.
37
“The Low-Income Home Energy Assistance Program: An Analysis of the 1984 Reauthorization Issues,” Coalition of
Northeastern Governors, April 1984, p. 9.
c11173008
Congressional Research Service
10
The LIHEAP Formula
.
Introduction of a Hold-Harmless Rate
After the House Energy and Commerce Committee reported H.R. 2439 to the House floor—but
before the full House could act on the bill—the Senate passed its version of LIHEAP
reauthorization as part of the Human Services Reauthorization Act (S. 2565) on October 4,
1984.
3838 The Senate bill contained language very similar to H.R. 2439, but made several changes
and additions to the formula.
•
S. 2565 specified that states
’' shares of LIHEAP funds would be based on the
home energy expenditures of low-income households, not on expenditures of all
households.
•
The hold-harmless level was altered. S. 2565 directed that no state in FY1985
would receive less funding than it received in FY1984, and for FY1986 and
thereafter, no state would receive less than the amount they would have received
in FY1984 if the appropriations level had been $1.975 billion.
•
A second hold-harmless provision, or hold-harmless rate, was created. The
provision maintained the
percentagepercentage allocated rather than a total funding level
allocated to each affected state.
The hold-harmless rate provision guaranteed that certain states would receive increased
allotments when appropriations reached $2.25 billion. States would qualify for this increase if
their total allotment percentage at an appropriation of $2.25 billion were less than 1%. These
states would instead receive the allotment rate they would have received at an appropriation of
$2.14 billion if that allotment rate were higher than the rate at $2.25 billion. In their debate about
S. 2565, Senators referred to the hold-harmless rate as the
“"small States hold harmless,
”" as the
intent was to protect the small (population) states
’' shares of LIHEAP funds.
3939 Otherwise, the
concern was that appropriations might have to increase significantly before small state allotments
would increase above their hold-harmless levels, with the states
’' percentage shares of funds
declining even as total appropriations increased.
The Senate bill also included different authorization amounts for LIHEAP, $2.14 billion for
FY1985 and $2.275 billion for FY1986. After S. 2565 passed the Senate, the House debated and
passed the bill on October 9, 1984, retaining all the provisions included in the Senate version. The
bill became P.L. 98-558, the Human Services Reauthorization Act, on October 30, 1984.
LIHEAP Formula Statutory Language
Unlike the allocation formulas under LIEAP and the other energy assistance programs that
preceded LIHEAP, which dictated the use of specific variables to determine allotments to the
states, the LIHEAP formula as drafted by Congress gives more general guidance to HHS. The
LIHEAP statute, as enacted in P.L. 98-558 and codified at 42 U.S.C. Section 8623(a)(2) provides
as follows.
38
The final version of S. 2565 can be found in the Congressional Record, daily edition, vol. 130 (October 4, 1984), p.
S13393.
39
Congressional Record, daily edition, vol. 130 (October 4, 1984), pp. S13415-S13416.
c11173008
Congressional Research Service
11
The LIHEAP Formula
.
as follows.
(A) a State
’'s allotment percentage is the percentage which expenditures for home energy by
low-income households in that State bears to such expenditures in all States, except that
States which thereby receive the greatest proportional increase in allotments by reason of the
application of this paragraph from the amount they received pursuant to P.L. 98-139 [the
FY1984 appropriation] shall have their allotments reduced to the extent necessary to ensure
that—
(i) no State for fiscal year 1985 shall receive less than the amount of funds the State
received in fiscal year 1984; and
(ii) no State for fiscal year 1986 and thereafter shall receive less than the amount of
funds the State would have received in fiscal year 1984 if the appropriations for this
subchapter for fiscal year 1984 had been $1,975,000,000, and
(B) any State whose allotment percentage out of funds available to States from a total
appropriation of $2,250,000,000 would be less than 1 percent, shall not, in any year when
total appropriations equal or exceed $2,250,000,000, have its allotment percentage reduced
from the percentage it would receive from a total appropriation of $2,140,000,000.
The next section of this report describes how funds are allocated to the states according to this
statutory language.
Determining LIHEAP Regular Fund Allotments
Using the
“New” Formula
"New" Formula
Current law as enacted in P.L. 98-558, sometimes referred to as the
“new”"new" LIHEAP formula,
provides for three different methods to calculate each state
’'s allotment of regular LIHEAP funds.
The calculation method used to determine state allotments depends upon the size of the
appropriation for that fiscal year.
•
If the annual appropriation level is at or below the equivalent of a hypothetical
FY1984 appropriation of $1.975 billion, then the allocation percentages under the
“old” "old" LIHEAP formula apply.
•
If appropriations exceed a hypothetical FY1984 appropriation of $1.975 billion,
then new formula percentages apply and are used to calculate state allotments. To
calculate the new formula percentages, HHS uses the most recent data available
to determine the heating and cooling costs of low-income households. When
appropriations exceed the $1.975 billion level, but are less than $2.25 billion, the
new formula percentages are used together with the hold-harmless level.
•
Finally, if appropriations equal or exceed $2.25 billion, the new percentages
apply and both the hold-harmless level together with the hold-harmless rate are in
effect.
This section describes the steps involved in allocating LIHEAP funds to the states under each of
the appropriations triggers.
c11173008
Congressional Research Service
12
The LIHEAP Formula
.
Calculating the New Formula Rates
As mentioned previously, when Congress considered a new formula for distributing LIHEAP
funds in 1983 and 1984, one of its concerns was the appropriateness and timeliness of the data
used in formula calculations. At the time, the energy information used to calculate state
allotments under the LIEAP formula did not use the most current data available.
4040 In fact, the
formula allocations were fixed percentages, and the LIHEAP statute at that time had no provision
for allowing newer information to be incorporated into the determination of state allotments. For
example, the formula used the change in cost of energy between 1978 and 1980, but did not take
account of increased costs after 1980. The LIHEAP formula as created by P.L. 98-558 requires
HHS to use the most recent data available. HHS updates these data periodically. The most recent
data were provided to CRS in
FebruaryNovember 2015.
As directed by the statute as enacted in 1984, the LIHEAP formula uses the home energy
expenditures of low-income households in each state as a first step in determining the proportion
of total regular funds that each state will receive.
4141 Specifically, this means estimating the amount
of money that all low-income households (as defined by the LIHEAP statute)
4242 in each state
spend on heating and cooling from all energy sources. This method accounts for variations in
heating and cooling needs of the states, the types of energy used, energy prices, and the
lowincomelow-income population and their heating and cooling methods. The process for capturing the
expenditures of low-income households for the most current year possible involves the following
steps.
•
Total Residential Energy Consumption. The first step in calculating new
formula rates is determining total residential energy consumption for each
heating and cooling source in every state. Residential energy consumption is
usually measured in terms of the total amount of British Thermal Units (Btus)
used in private households and generally captures energy used for space and
water heating, cooling, lighting, refrigeration, cooking, and the energy needed to
operate appliances. The most recent data used in calculating LIHEAP formula
rates come from the
20122013 Energy Information Administration (EIA) State Energy
Data System consumption estimates.
•
Temperature Variation. The next step in determining the formula rates involves
adjusting the amount of energy consumed for each fuel source by temperature
variation in each state. This is done by using a ratio consisting of the 30-year
average heating and cooling degree day data to each state
’'s share of the most
recent year
’'s average heating and cooling degree days. A heating degree day
measures the extent to which a day
’'s average temperature falls below 65°F and a
cooling degree day measures the extent to which a day
’'s average temperature
rises above 65°F.
4343 For example, a day with an average temperature of 50°F
results in a measure of 15 heating degree days; a day with an average temperature
40
Report of the Committee on Energy and Commerce to accompany H.R. 2439, the Low-Income Home Energy
Amendments of 1984, 98th Cong., 2nd sess., H.Rept. 98-139, Part 2, May 15, 1984, p. 13.
41
“[A] State’s allotment percentage is the percentage which expenditures for home energy by low-income households
in that State bears to such expenditures in all States.” 42 U.S.C. §8623(a)(2).
42
The LIHEAP statute considers households with income at or below 150% of poverty or 60% of state median income
(whichever value is greater) to be low income. 42 U.S.C. §8624(b)(2)(B).
43
A state’s heating and cooling degree data are weighted by population in the state.
c11173008
Congressional Research Service
13
The LIHEAP Formula
.
results in a measure of 15 heating degree days; a day with an average temperature of 80°F results in a measure of 15 cooling degree days. The purpose of the
adjustment to fuel consumption is to account for abnormally warm or cool years,
where energy usage might attain extreme values. This information is collected by
the National Oceanic and Atmospheric Administration. The most recent year
’s
's average heating and cooling degree day data are from
20122013, and the 30-year
average was computed from 1971 to 2000.
44
45
c11173008
•
Heating and Cooling Consumption. As mentioned above, total residential
energy consumption encompasses other uses in addition to heating and cooling
(e.g., operation of appliances). So the next step in calculating LIHEAP formula
rates is to derive the portion of fuel consumed specifically to heat and cool homes
as opposed to other uses. The EIA, as part of the Residential Energy
Consumption Survey (RECS), uses an
“"end use estimation methodology
”" to
estimate the amount of fuel used for heating and cooling (among other uses). The
most recent information on heating and cooling consumption comes from the
2009 RECS.
4444 HHS adjusts the EIA heating and cooling consumption estimates
using heating degree day and cooling degree day data.
•
Low-Income Household Heating and Cooling Consumption. After estimating
heating and cooling consumption for
allall households, the next step is to calculate
heating and cooling consumption in Btus for low-income households. HHS uses
Census data to determine fuel sources used by low-income households. The most
recent information on low-income households and the fuel sources they use
comes from the American Community Survey three-year estimates for
201020122011-2013. In addition, low-income consumption data are adjusted to account for the
fact that low-income households might use more or less of a fuel source than is
used by households on average. This is done using consumption data from the
2009 RECS.
•
Total Spending on Heating and Cooling. To arrive at the amount of money that
low-income households spend on heating and cooling, the number of Btus used
by low-income households that were estimated in the previous step are multiplied
by the average fuel price for each fuel source. The total amount spent on heating
and cooling by low-income households for each fuel source is then added
together to arrive at total spending for each state. Regional energy price variation
can be significant, and the formula takes expected expenditure differences into
account. This information is collected by the EIA and published in the State
Energy Data System Consumption, Price, and Expenditure Estimates.
4545 The most
recent price data used to calculate formula rates are from
2012.
•
2013.
New Formula Percentage. Finally, these expenditure data are used to estimate
the amount spent by low-income households on heating and cooling in each state
relative to the amount spent by low-income households on heating and cooling in
all states. The calculated proportion becomes the new formula percentage for
each state. Table 3
at the end of this section shows both the percentages under
the “old” the "old" formula (column (a)) and the most recent
“new”"new" formula percentages
(column (b)), received by CRS from HHS in
FebruaryNovember 2015. To see how the
formula rates for each state have changed in recent years, see
Table 4.Table 4.
For more information about the RECS, see the EIA website at http://www.eia.doe.gov/emeu/recs/.
The EIA’s state data tables are available at http://www.eia.doe.gov/emeu/states/_seds.html.
Congressional Research Service
14
The LIHEAP Formula
.
These new formula percentages are used to allocate LIHEAP funds to the states if the annual
appropriation exceeds the equivalent of a hypothetical FY1984 appropriation of $1.975 billion.
However, they do not represent the exact percentage of funds that states will receive under the
new formula. The ultimate allotments are determined after application of both the hold-harmless
level and hold-harmless rate, described in the next section. The new percentages are the starting
point for determining how funds will be allocated to the states.
Using the New Formula Percentages to Allocate Funds to the States
The LIHEAP new formula percentages that HHS calculates using the most current data available
do not necessarily represent the percentage of funds that states will receive. State allotments
depend upon the application of the two hold-harmless provisions in the LIHEAP statute. Some
states must have their share of funds ratably reduced in order to hold harmless those states that
would, but for the hold-harmless provisions, lose funds. Other states see a gain in their share of
funds because they benefit from the hold-harmless provisions. The application of the
holdharmlesshold-harmless provisions depends upon the size of the appropriation for a given fiscal year. These
appropriation level triggers are described below.
“Old”
"Old" Formula: Appropriations At or Below $1.975 Billion
The LIHEAP statute does not contain an explicit trigger for the
“new”"new" formula rates to be used.
However, the statute specifies that states must receive no less than
“"the amount of funds the State
would have received in fiscal year 1984 if the appropriations for this subchapter for fiscal year
1984 had been $1,975,000,000.
”" As a result, up to this appropriation level, states receive the same
percentage of funds that they would have received at a given appropriation level under the
“old”
LIHEAP formula.46
"old" LIHEAP formula.46
The FY1984 appropriation of $1.975 billion referred to in the LIHEAP statute is hypothetical
because this was not the amount actually appropriated in FY1984. The actual FY1984
appropriation was $2.075 billion. In addition, the current year appropriation that is
“equivalent
to”"equivalent to" a hypothetical FY1984 appropriation of $1.975 billion is not exactly $1.975 billion. In
FY1984, with the exception of funds provided to the territories, all LIHEAP regular funds were
distributed to the states. Since then, two other funds have become part of the regular fund
distribution. These are funds for training and technical assistance and for the leveraging incentive
grants (which includes REACH grants) to the states. This means that an appropriation that is
equivalent to a hypothetical FY1984 appropriation of $1.975 billion must account for these new
funds. Assuming that funds for leveraging incentive/REACH grants would be $27 million and
training and technical assistance would be $3 million (amounts that have typically been set aside
in the appropriation), then the equivalent of an FY1984 appropriation of $1.975 billion is
approximately $2.005 billion.47
46
When appropriations are below a hypothetical FY1984 appropriation of $1.975 billion, the result of the current law’s
hold-harmless provisions is that states receive the same allotment percentages that they did under the old formula. See
U.S. Department of Health and Human Services, Low Income Home Energy Assistance Program: Report to Congress
for FY1987, p. 133.
47
This amount is arrived at by adding $27 million and $3 million to $1.975 billion.
c11173008
Congressional Research Service
15
The LIHEAP Formula
.
approximately $2.005 billion.47
The LIHEAP formula in FY1984 distributed funds by giving states the same percentage of funds
that they received in FY1981 under the predecessor program, the Low Income Energy Assistance
Program (LIEAP). Table 3 (later, following the
“"Implementation of the
“New” LIHEAP
Formula”"New" LIHEAP Formula" section), shows rates under the old formula in column (a). For example, at an
appropriation at or below the equivalent of a hypothetical FY1984 appropriation of $1.975
billion, Alabama would receive 0.86% of total funds, Alaska would receive 0.55% of total funds,
and so on. Table A-1, column (a) reports the dollar amount of funds that each state would have
received in FY1984 had the regular fund appropriation been $1.975 billion. For comparison
purposes, the dollar amounts also assume that funds for the territories would be 0.5% of the total,
a change made by HHS beginning with the FY2014 appropriation.
48
“New”48
"New" Formula with Hold-Harmless Level: Appropriations Between $1.975
Billion and $2.25 Billion
If the regular LIHEAP appropriation exceeds the equivalent of a hypothetical FY1984
appropriation of $1.975 billion for the fiscal year,
allall funds are to be distributed under a different
methodology, using the new set of percentages described earlier. In addition, a hold-harmless
level level applies to ensure that certain states do not fall below the amount of funds they would have
received at the equivalent of a hypothetical FY1984 appropriation of $1.975 billion.
Table 3 shows whether a state benefits from the hold-harmless level. This is indicated by a
“Y” in
"Y" in column (c), while the dollar amount of funds those states receive by being held harmless appears
in column (d). For example, Alabama is not held harmless, while Colorado is held harmless. The
dollar amount of funds that Colorado receives pursuant to the hold-harmless level is $31.613
million. But for the hold-harmless level, Colorado would receive less than this dollar amount at
its new formula percentage at certain appropriation levels. Eventually, when appropriations
increase sufficiently, the percentage of funds under the
“new”"new" formula for hold-harmless states
will exceed their hold harmless amounts and they will begin to receive their
“new”"new" percentage of
funds. This appropriation level varies for each state. For example, at lower appropriation levels,
the $31.613 million hold-harmless level for Colorado exceeds the state
’s “new”'s "new" percentage share
of 1.
391413% of total funds. However, at an appropriation of about $2.3 billion, Colorado
’'s new
percentage share exceeds $31.613 million and the state begins to receive funds at the
“new”
"new" percentage. Eventually, many states will receive the percentage of funds at their
“new”
percentage.49
"new" percentage.49
The hold-harmless level is achieved by reducing the allocation of funds to states with the greatest
proportional gains under the new formula percentages.
5050 For example, under the most recent
LIHEAP formula percentages, states with the greatest proportional gains were Nevada, Arizona,
and Florida and Texas. Depending on the appropriation level, these states (and others with the greatest
gains) may then have their allotments reduced to hold harmless the states that would otherwise
48
HHS Administration for Children and Families, Office of Community Services, LIHEAP Dear Colleague Notice
Allocation for Territories FY2014, November 22, 2013, http://www.acf.hhs.gov/programs/ocs/resource/liheapallocation-for-territories-fy-2014.
49
The exceptions to this are states that benefit from the hold-harmless rate, described in the next section, and the states
that are ratably reduced in order to compensate states that benefit from the hold-harmless rate.
50
“States which thereby receive the greatest proportional increase in allotments ... shall have their allotments reduced
to the extent necessary to ensure that ... no State for fiscal year 1986 and thereafter shall receive less than the amount of
funds the State would have received in fiscal year 1984.” 42 U.S.C. §8623(a)(2)(A)(ii).
c11173008
Congressional Research Service
16
The LIHEAP Formula
.
see reduced benefits. So although these states with the greatest proportional gains will see their
LIHEAP allotments increase under the new formula, their allotments may not increase to reach
their new formula rates (column (b) of Table 3
).
).
Columns (b) and (c) of Table A-1 show estimated allotments to the states at hypothetical
appropriations levels between $1.975 billion and $2.25 billion. Column (b) shows the estimated
allotment of funds that each state would receive when the regular fund appropriation is at $2.14
billion and column (c) shows the estimated allotment of funds when the regular fund
appropriation is just under $2.25 billion ($2,249,999,999).
“New”
"New" Formula with Hold-Harmless Level and Rate: Appropriations At or
Above $2.25 Billion
The LIHEAP statute stipulates additional requirements in the method for distributing funds when
the appropriation is at or above $2.25 billion. At this level, the hold-harmless level still applies,
but, in addition, a new hold-harmless
raterate is applied. Specifically, for all appropriation levels at or
above $2.25 billion, states that would have received less than 1% of a total $2.25 billion
appropriation must be allocated the percentage they would have received at a $2.14 billion
appropriation level.
5151 (This assumes the percentage at $2.14 billion is greater than the percentage
originally calculated at the hypothetical $2.25 billion appropriation; this is not true for all states
that receive less than 1% of the $2.25 billion appropriation.) Then that state will receive the
percentage share of funds it would have received at $2.14 billion for all appropriation levels at or
above $2.25 billion. This hold-harmless
raterate ensures a state specific
shareshare of the total available
funds.
As with the hold-harmless level, the allocations to the states with the greatest proportional gains
are then ratably reduced again until there is no funding shortfall. Column (e) of Table 3 shows
which states benefit from the hold-harmless rate, indicated by a
“Y,”"Y," while column (f) shows the
proportion of funds that those states receive. For example, Idaho benefits from the hold-harmless
rate and receives 0.587% of the total appropriation when appropriations are at or above $2.25
billion.
The application of the hold-harmless rate creates another layer of discontinuity in the allocation
rates. States that are ratably reduced see their allocations at $2.25 billion fall below the amount
they would receive at $2.249 billion, while states that benefit from the hold-harmless rate see
their funding jump up slightly. Columns (d) through (i) of Table A-1 in Appendix A show
estimated allotments to states at various hypothetical appropriations levels at or above $2.25
billion. Column (d) shows the estimated allotment of funds that each state receives when the
regular appropriation is at $2.25 billion after the hold-harmless rate is applied. Columns (e)
through (i) show the estimated allotment each state would receive at $2.5 billion, $3.0 billion,
$3.39 billion (the amount appropriated in FY2014 and FY2015), $4.0 billion, and $5.1 billion
(the amount at which LIHEAP was last authorized).
51
“[A]ny State whose allotment percentage out of funds available to States from a total appropriation of
$2,250,000,000 would be less than 1 percent, shall not, in any year when total appropriations equal or exceed
$2,250,000,000, have its allotment percentage reduced from the percentage it would receive from a total appropriation
of $2,140,000,000.” 42 U.S.C. §8623(a)(2)(B).
c11173008
Congressional Research Service
17
The LIHEAP Formula
.
(the amount at which LIHEAP was last authorized).
Implementation of the
“New”"New" LIHEAP Formula
Until FY2006, appropriations for regular LIHEAP funds had only exceeded the equivalent of a
hypothetical FY1984 appropriation of $1.975 billion in 1985 and 1986; thereafter, from FY1987
through FY2005, and again in FY2007, states continued to receive the same percentage of
LIHEAP funds that they received under the program
’'s predecessor, LIEAP (see column (a) of
Table 3 for these percentages). In FY2006, funds were distributed under the
“new”"new" LIHEAP
formula when Congress appropriated $2.48 billion in regular funds for the program. In FY2008,
perhaps due to an oversight, the new formula was again used to distribute funds. The FY2008
Consolidated Appropriations Act (P.L. 110-161) failed to authorize a set-aside called leveraging
incentive grants. As a result, the funds for those grants were added to the LIHEAP regular funds,
triggering the new formula.
5252 In FY2009, the Consolidated Security, Disaster Assistance, and
Continuing Appropriations Act (P.L. 110-329) appropriated $4.51 billion in regular funds.
However, the law further specified that $840 million be distributed according to the
“new”
"new" LIHEAP formula, with the remaining $3.67 billion distributed according to the percentages of the
“old” "old" formula established by LIEAP. From FY2010 to
FY2015FY2016, Congress has continued to
appropriate funds using a version of a split between the
“old” and “new”"old" and "new" formulas. See Table C-1
in Appendix C of this report for the distribution of funds to the states from FY2009 through
FY2015.
Table 3. Low-Income Home Energy Program (LIHEAP):
“Old” and “New” Allotment Percentages by State, FY2015
Hold-Harmless Levela
State
“New”
Allotment
Percentage
(%)
(b)
State Held
Harmless?
(c)
HoldHarmless
Level
($Millions)
(d)
State Held
Harmless?
(e)
HoldHarmless
Rate (%)
(f)
Alabama
0.860
1.488
N
—
N
—
Alaska
0.549
0.491
Y
10.788
Y
0.514
Arizona
0.416
1.424
N
—
N
—
Arkansas
0.656
0.846
N
—
N
—
California
4.614
5.371
N
—
N
—
Colorado
1.609
1.391
Y
31.613
N
—
Connecticut
2.099
2.711
N
—
N
—
Delaware
0.279
0.407
N
—
N
—
District of
Columbia
0.326
0.173
Y
6.405
Y
0.305
Florida
1.361
4.057
N
—
N
—
Georgia
1.076
3.068
N
—
N
—
Hawaii
0.108
0.219
N
—
N
—
Idaho
0.628
0.364
Y
12.331
Y
0.587
52
c11173008
“Old”
Allotment
Percentage
(%)
(a)
Hold-Harmless Rate
For more information about this issue, see Appendix C of this report.
Congressional Research Service
18
The LIHEAP Formula
.
Hold-Harmless Levela
State
“New”
Allotment
Percentage
(%)
(b)
State Held
Harmless?
(c)
HoldHarmless
Level
($Millions)
(d)
State Held
Harmless?
(e)
HoldHarmless
Rate (%)
(f)
Illinois
5.809
4.075
Y
114.147
N
—
Indiana
2.630
1.712
Y
51.683
N
—
Iowa
1.864
1.005
Y
36.628
N
—
Kansas
0.856
0.932
N
—
N
—
Kentucky
1.369
1.318
Y
26.895
N
—
Louisiana
0.879
1.236
N
—
N
—
Maine
1.360
1.052
Y
26.717
N
—
Maryland
1.607
2.206
N
—
N
—
Massachusetts
4.198
4.395
N
—
N
—
Michigan
5.515
4.535
Y
108.373
N
—
Minnesota
3.973
1.827
Y
78.076
N
—
Mississippi
0.737
0.825
N
—
N
—
Missouri
2.320
2.140
Y
45.595
N
—
Montana
0.736
0.347
Y
14.464
Y
0.689
Nebraska
0.922
0.483
Y
18.114
Y
0.863
Nevada
0.195
0.722
N
—
N
—
New
Hampshire
0.795
0.753
Y
15.615
N
—
New Jersey
3.897
3.703
Y
76.584
N
—
New Mexico
0.521
0.533
N
—
N
—
12.725
10.792
Y
250.058
N
—
North
Carolina
1.896
2.817
N
—
N
—
North Dakota
0.800
0.251
Y
15.712
Y
0.748
Ohio
5.139
3.836
Y
100.980
N
—
Oklahoma
0.791
1.186
N
—
N
—
Oregon
1.247
0.885
Y
24.502
N
—
Pennsylvania
6.835
5.856
Y
134.318
N
—
Rhode Island
0.691
0.857
N
—
N
—
South
Carolina
0.683
1.288
N
—
N
—
South Dakota
0.649
0.242
Y
12.761
Y
0.608
Tennessee
1.386
1.730
N
—
N
—
Texas
2.264
6.529
N
—
N
—
New York
c11173008
“Old”
Allotment
Percentage
(%)
(a)
Hold-Harmless Rate
Congressional Research Service
19
The LIHEAP Formula
.
Hold-Harmless Levela
State
“Old”
Allotment
Percentage
(%)
(a)
“New”
Allotment
Percentage
(%)
(b)
Hold-Harmless Rate
State Held
Harmless?
(c)
HoldHarmless
Level
($Millions)
(d)
State Held
Harmless?
(e)
HoldHarmless
Rate (%)
(f)
Utah
0.748
0.568
Y
14.691
Y
0.700
Vermont
0.596
0.490
Y
11.704
Y
0.557
Virginia
1.957
2.588
N
—
N
—
Washington
2.051
1.443
Y
40.302
N
—
West Virginia
0.906
0.661
Y
17.799
Y
0.848
Wisconsin
3.576
2.000
Y
70.280
N
—
Wyoming
0.299
0.173
Y
5.882
Y
0.280
Source: New allotment percentages were provided to CRS by HHS in February 2015. Information in columns
(c) through (f) are based on CRS calculations using the new allotment percentages. The calculations assume that
funding would be provided for leveraging incentive/REACH grants, training and technical assistance, and 0.5% for
the territories.
Notes: The actual percentage of total regular funds each state receives at funding levels above $1.975 billion
may differ from the new formula percentages due to the hold-harmless provisions and the ratable reductions of
some states to cover shortfall from these hold-harmless provisions.
a.
The states that benefit from the hold-harmless level vary depending on the amount appropriated for
LIHEAP regular funds. The states listed here benefit from the hold-harmless level when appropriations just
exceed the equivalent of an FY1984 appropriation of $1.975 billion.
Table 4. Recent State Allotment Percentages Under the “New” LIHEAP Formula
(Fiscal years indicate when new formula rates would have been used to distribute funds to states)
States
“Old”
Formula
Percentages
FY2008
FY2009
FY2010
FY2011
FY2012
FY2013
FY2014
FY2015
Alabama
0.860%
1.932%
1.650%
1.582%
1.599%
1.583%
1.716%
1.686%
1.488%
Alaska
0.549
0.376
0.317
0.575
0.511
0.398
0.522
0.563
0.491
Arizona
0.416
0.992
0.813
1.018
1.098
1.132
1.326
1.379
1.424
Arkansas
0.656
1.082
0.910
0.884
0.852
0.899
0.876
0.876
0.846
California
4.614
5.690
5.303
4.479
4.453
4.452
4.433
4.536
5.371
Colorado
1.609
1.280
1.305
1.333
1.247
1.267
1.264
1.270
1.391
Connecticut
2.099
1.732
2.164
2.205
2.239
2.398
2.416
2.371
2.711
Delaware
0.279
0.435
0.453
0.375
0.373
0.375
0.421
0.427
0.407
District of
Columbia
0.326
0.309
0.328
0.181
0.192
0.194
0.184
0.149
0.173
Florida
1.361
4.187
3.781
4.728
4.583
4.593
5.475
5.201
4.057
Georgia
1.076
2.829
2.734
2.620
2.641
2.742
3.137
3.166
3.068
Hawaii
0.108
0.101
0.099
0.150
0.150
0.205
0.185
0.230
0.219
Idaho
0.628
0.386
0.331
0.396
0.349
0.335
0.339
0.371
0.364
c11173008
“New” Formula Percentages
Congressional Research Service
20
The LIHEAP Formula
.
States
“Old”
Formula
Percentages
FY2008
FY2009
FY2010
FY2011
FY2012
FY2013
FY2014
FY2015
Illinois
5.809
4.796
4.998
4.843
5.014
5.243
4.655
4.510
4.075
Indiana
2.630
2.209
2.128
2.147
2.080
2.209
1.814
1.934
1.712
Iowa
1.864
1.085
1.064
1.028
1.099
1.080
1.001
1.065
1.005
Kansas
0.856
1.105
1.106
0.978
0.993
0.967
1.002
0.945
0.932
Kentucky
1.369
1.688
1.621
1.243
1.256
1.344
1.329
1.457
1.318
Louisiana
0.879
1.704
1.514
1.324
1.365
1.414
1.378
1.387
1.236
Maine
1.360
0.722
0.908
1.127
1.090
1.010
0.927
1.041
1.052
Maryland
1.607
2.421
2.652
1.965
2.080
2.197
2.344
2.193
2.206
Massachusetts
4.198
3.043
3.311
3.757
3.718
3.730
4.032
4.138
4.395
Michigan
5.515
4.651
4.645
5.040
4.819
4.863
4.966
4.681
4.535
Minnesota
3.973
1.789
1.917
2.023
2.025
2.047
1.849
1.921
1.827
Mississippi
0.737
1.105
0.951
0.974
0.940
0.990
0.955
0.953
0.825
Missouri
2.320
2.497
2.309
2.014
2.011
1.829
1.963
2.021
2.140
Montana
0.736
0.414
0.441
0.295
0.287
0.328
0.280
0.314
0.347
Nebraska
0.922
0.598
0.558
0.547
0.553
0.591
0.555
0.561
0.483
Nevada
0.195
0.686
0.576
0.500
0.526
0.498
0.563
0.537
0.722
New
Hampshire
0.795
0.453
0.503
0.612
0.605
0.742
0.623
0.731
0.753
New Jersey
3.897
2.838
3.621
3.995
4.105
4.010
3.812
3.620
3.703
New Mexico
0.521
0.628
0.577
0.458
0.441
0.430
0.407
0.394
0.533
New York
12.725
8.491
9.393
9.520
10.018
10.227
9.445
9.318
10.792
North
Carolina
1.896
3.186
3.261
2.766
2.823
2.619
2.954
2.891
2.817
North Dakota
0.800
0.235
0.273
0.246
0.256
0.302
0.215
0.254
0.251
Ohio
5.139
4.512
4.803
4.893
4.941
4.687
4.243
4.368
3.836
Oklahoma
0.791
1.452
1.275
1.236
1.224
1.152
1.207
1.219
1.186
Oregon
1.247
1.008
0.750
0.715
0.702
0.664
0.712
0.781
0.885
Pennsylvania
6.835
5.174
5.731
5.993
5.885
5.807
5.571
5.720
5.856
Rhode Island
0.691
0.596
0.665
0.635
0.615
0.670
0.753
0.712
0.857
South
Carolina
0.683
1.425
1.349
1.278
1.260
1.201
1.394
1.403
1.288
South Dakota
0.649
0.268
0.235
0.249
0.253
0.272
0.233
0.240
0.242
Tennessee
1.386
2.055
1.801
1.743
1.717
1.700
1.865
1.848
1.730
Texas
2.264
7.095
6.524
7.668
7.349
7.135
7.183
6.942
6.529
Utah
0.748
0.648
0.599
0.559
0.508
0.413
0.452
0.494
0.568
Vermont
0.596
0.356
0.319
0.418
0.419
0.396
0.417
0.425
0.490
Virginia
1.957
2.817
3.041
2.428
2.486
2.490
2.581
2.607
2.588
c11173008
“New” Formula Percentages
Congressional Research Service
21
The LIHEAP Formula
.
“Old”
Formula
Percentages
FY2008
FY2009
FY2010
FY2011
FY2012
FY2013
FY2014
FY2015
Washington
2.051
1.621
1.204
1.225
1.245
1.145
1.244
1.305
1.443
West Virginia
0.906
0.960
0.907
0.663
0.639
0.638
0.625
0.631
0.661
Wisconsin
3.576
2.108
2.080
2.229
2.236
2.230
2.010
2.054
2.000
Wyoming
0.299
0.233
0.202
0.137
0.129
0.154
0.146
0.160
0.173
States
“New” Formula Percentages
Source: State data were received by CRS from HHS in May 2007, September 2008, April 2009, June 2010,
August 2011, August 2012, September 2013, and February 2015.
c11173008
Congressional Research Service
22
The LIHEAP Formula
.
Appendix A. Estimated Allotments to the States
Under Various Hypothetical Appropriations Levels
Table A-1, below, shows estimated allocations to the states at various hypothetical appropriations
levels. In column (a) are allotments at the equivalent of a hypothetical FY1984 appropriation of
$1.975 billion—under current LIHEAP practice where funds are set aside for leveraging incentive
grants and training and technical assistance, the equivalent appropriation level is approximately
$2.005 billion. The remaining columns show estimated allotments at appropriations of $2.14
billion, just under $2.25 billion, $2.25 billion, $3.0 billion, $3.39 billion, $4.0 billion, and $5.1
FY2015, and estimated allocations for FY2016.
Table 3. Low-Income Home Energy Program (LIHEAP): "Old" and "New" Allotment Percentages by State, FY2016
Hold-Harmless Levela
Hold-Harmless Rate
|
State
|
"Old" Allotment Percentage (%) (a)
"New" Allotment Percentage (%) (b)
State Held Harmless? (c)
Hold-Harmless Level ($Millions) (d)
State Held Harmless? (e)
Hold-Harmless Rate (%) (f)
Alabama
|
0.860
|
1.521
|
N
|
—
|
N
|
—
|
Alaska
|
0.549
|
0.432
|
Y
|
10.788
|
Y
|
0.514
|
Arizona
|
0.416
|
1.440
|
N
|
—
|
N
|
—
|
Arkansas
|
0.656
|
0.891
|
N
|
—
|
N
|
—
|
California
|
4.614
|
5.504
|
N
|
—
|
N
|
—
|
Colorado
|
1.609
|
1.413
|
Y
|
31.613
|
N
|
—
|
Connecticut
|
2.099
|
2.508
|
N
|
—
|
N
|
—
|
Delaware
|
0.279
|
0.409
|
N
|
—
|
N
|
—
|
District of Columbia
|
0.326
|
0.189
|
Y
|
6.405
|
Y
|
0.305
|
Florida
|
1.361
|
3.936
|
N
|
—
|
N
|
—
|
Georgia
|
1.076
|
2.924
|
N
|
—
|
N
|
—
|
Hawaii
|
0.108
|
0.196
|
N
|
—
|
N
|
—
|
Idaho
|
0.628
|
0.387
|
Y
|
12.331
|
Y
|
0.587
|
Illinois
|
5.809
|
4.245
|
Y
|
114.147
|
N
|
—
|
Indiana
|
2.630
|
1.792
|
Y
|
51.683
|
N
|
—
|
Iowa
|
1.864
|
1.054
|
Y
|
36.628
|
N
|
—
|
Kansas
|
0.856
|
0.982
|
N
|
—
|
N
|
—
|
Kentucky
|
1.369
|
1.395
|
N
|
—
|
N
|
—
|
Louisiana
|
0.879
|
1.394
|
N
|
—
|
N
|
—
|
Maine
|
1.360
|
1.066
|
Y
|
26.717
|
N
|
—
|
Maryland
|
1.607
|
2.347
|
N
|
—
|
N
|
—
|
Massachusetts
|
4.198
|
4.501
|
N
|
—
|
N
|
—
|
Michigan
|
5.515
|
4.357
|
Y
|
108.373
|
N
|
—
|
Minnesota
|
3.973
|
1.869
|
Y
|
78.076
|
N
|
—
|
Mississippi
|
0.737
|
0.910
|
N
|
—
|
N
|
—
|
Missouri
|
2.320
|
2.145
|
Y
|
45.595
|
N
|
—
|
Montana
|
0.736
|
0.358
|
Y
|
14.464
|
Y
|
0.689
|
Nebraska
|
0.922
|
0.531
|
Y
|
18.114
|
Y
|
0.863
|
Nevada
|
0.195
|
0.718
|
N
|
—
|
N
|
—
|
New Hampshire
|
0.795
|
0.788
|
Y
|
15.615
|
N
|
—
|
New Jersey
|
3.897
|
3.766
|
Y
|
76.584
|
N
|
—
|
New Mexico
|
0.521
|
0.571
|
N
|
—
|
N
|
—
|
New York
|
12.725
|
9.477
|
Y
|
250.058
|
N
|
—
|
North Carolina
|
1.896
|
2.828
|
N
|
—
|
N
|
—
|
North Dakota
|
0.800
|
0.275
|
Y
|
15.712
|
Y
|
0.748
|
Ohio
|
5.139
|
3.850
|
Y
|
100.980
|
N
|
—
|
Oklahoma
|
0.791
|
1.207
|
N
|
—
|
N
|
—
|
Oregon
|
1.247
|
0.860
|
Y
|
24.502
|
N
|
—
|
Pennsylvania
|
6.835
|
5.810
|
Y
|
134.318
|
N
|
—
|
Rhode Island
|
0.691
|
0.803
|
N
|
—
|
N
|
—
|
South Carolina
|
0.683
|
1.323
|
N
|
—
|
N
|
—
|
South Dakota
|
0.649
|
0.257
|
Y
|
12.761
|
Y
|
0.608
|
Tennessee
|
1.386
|
1.771
|
N
|
—
|
N
|
—
|
Texas
|
2.264
|
6.870
|
N
|
—
|
N
|
—
|
Utah
|
0.748
|
0.526
|
Y
|
14.691
|
Y
|
0.700
|
Vermont
|
0.596
|
0.503
|
Y
|
11.704
|
Y
|
0.557
|
Virginia
|
1.957
|
2.692
|
N
|
—
|
N
|
—
|
Washington
|
2.051
|
1.434
|
Y
|
40.302
|
N
|
—
|
West Virginia
|
0.906
|
0.707
|
Y
|
17.799
|
Y
|
0.848
|
Wisconsin
|
3.576
|
2.100
|
Y
|
70.280
|
N
|
—
|
Wyoming
|
0.299
|
0.169
|
Y
|
5.882
|
Y
|
0.280
|
Source: New allotment percentages were provided to CRS by HHS in November 2015. Information in columns (c) through (f) are based on CRS calculations using the new allotment percentages. The calculations assume that funding would be provided for leveraging incentive/REACH grants, training and technical assistance, and 0.5% for the territories.
Notes: The actual percentage of total regular funds each state receives at funding levels above $1.975 billion may differ from the new formula percentages due to the hold-harmless provisions and the ratable reductions of some states to cover shortfall from these hold-harmless provisions.
a.
The states that benefit from the hold-harmless level vary depending on the amount appropriated for LIHEAP regular funds. The states listed here benefit from the hold-harmless level when appropriations just exceed the equivalent of an FY1984 appropriation of $1.975 billion.
Table 4. Recent State Allotment Percentages Under the "New" LIHEAP Formula
(Fiscal years indicate when new formula rates would have been used to distribute funds to states)
"New" Formula Percentages
|
States
|
"Old" Formula Percentages
|
FY2009
|
FY2010
|
FY2011
|
FY2012
|
FY2013
|
FY2014
|
FY2015
|
FY2016
|
Alabama
|
0.860%
|
1.650%
|
1.582%
|
1.599%
|
1.583%
|
1.716%
|
1.686%
|
1.488%
|
1.521%
|
Alaska
|
0.549
|
0.317
|
0.575
|
0.511
|
0.398
|
0.522
|
0.563
|
0.491
|
0.432
|
Arizona
|
0.416
|
0.813
|
1.018
|
1.098
|
1.132
|
1.326
|
1.379
|
1.424
|
1.440
|
Arkansas
|
0.656
|
0.910
|
0.884
|
0.852
|
0.899
|
0.876
|
0.876
|
0.846
|
0.891
|
California
|
4.614
|
5.303
|
4.479
|
4.453
|
4.452
|
4.433
|
4.536
|
5.371
|
5.504
|
Colorado
|
1.609
|
1.305
|
1.333
|
1.247
|
1.267
|
1.264
|
1.270
|
1.391
|
1.413
|
Connecticut
|
2.099
|
2.164
|
2.205
|
2.239
|
2.398
|
2.416
|
2.371
|
2.711
|
2.508
|
Delaware
|
0.279
|
0.453
|
0.375
|
0.373
|
0.375
|
0.421
|
0.427
|
0.407
|
0.409
|
District of Columbia
|
0.326
|
0.328
|
0.181
|
0.192
|
0.194
|
0.184
|
0.149
|
0.173
|
0.189
|
Florida
|
1.361
|
3.781
|
4.728
|
4.583
|
4.593
|
5.475
|
5.201
|
4.057
|
3.936
|
Georgia
|
1.076
|
2.734
|
2.620
|
2.641
|
2.742
|
3.137
|
3.166
|
3.068
|
2.924
|
Hawaii
|
0.108
|
0.099
|
0.150
|
0.150
|
0.205
|
0.185
|
0.230
|
0.219
|
0.196
|
Idaho
|
0.628
|
0.331
|
0.396
|
0.349
|
0.335
|
0.339
|
0.371
|
0.364
|
0.387
|
Illinois
|
5.809
|
4.998
|
4.843
|
5.014
|
5.243
|
4.655
|
4.510
|
4.075
|
4.245
|
Indiana
|
2.630
|
2.128
|
2.147
|
2.080
|
2.209
|
1.814
|
1.934
|
1.712
|
1.792
|
Iowa
|
1.864
|
1.064
|
1.028
|
1.099
|
1.080
|
1.001
|
1.065
|
1.005
|
1.054
|
Kansas
|
0.856
|
1.106
|
0.978
|
0.993
|
0.967
|
1.002
|
0.945
|
0.932
|
0.982
|
Kentucky
|
1.369
|
1.621
|
1.243
|
1.256
|
1.344
|
1.329
|
1.457
|
1.318
|
1.395
|
Louisiana
|
0.879
|
1.514
|
1.324
|
1.365
|
1.414
|
1.378
|
1.387
|
1.236
|
1.394
|
Maine
|
1.360
|
0.908
|
1.127
|
1.090
|
1.010
|
0.927
|
1.041
|
1.052
|
1.066
|
Maryland
|
1.607
|
2.652
|
1.965
|
2.080
|
2.197
|
2.344
|
2.193
|
2.206
|
2.347
|
Massachusetts
|
4.198
|
3.311
|
3.757
|
3.718
|
3.730
|
4.032
|
4.138
|
4.395
|
4.501
|
Michigan
|
5.515
|
4.645
|
5.040
|
4.819
|
4.863
|
4.966
|
4.681
|
4.535
|
4.357
|
Minnesota
|
3.973
|
1.917
|
2.023
|
2.025
|
2.047
|
1.849
|
1.921
|
1.827
|
1.869
|
Mississippi
|
0.737
|
0.951
|
0.974
|
0.940
|
0.990
|
0.955
|
0.953
|
0.825
|
0.910
|
Missouri
|
2.320
|
2.309
|
2.014
|
2.011
|
1.829
|
1.963
|
2.021
|
2.140
|
2.145
|
Montana
|
0.736
|
0.441
|
0.295
|
0.287
|
0.328
|
0.280
|
0.314
|
0.347
|
0.358
|
Nebraska
|
0.922
|
0.558
|
0.547
|
0.553
|
0.591
|
0.555
|
0.561
|
0.483
|
0.531
|
Nevada
|
0.195
|
0.576
|
0.500
|
0.526
|
0.498
|
0.563
|
0.537
|
0.722
|
0.718
|
New Hampshire
|
0.795
|
0.503
|
0.612
|
0.605
|
0.742
|
0.623
|
0.731
|
0.753
|
0.788
|
New Jersey
|
3.897
|
3.621
|
3.995
|
4.105
|
4.010
|
3.812
|
3.620
|
3.703
|
3.766
|
New Mexico
|
0.521
|
0.577
|
0.458
|
0.441
|
0.430
|
0.407
|
0.394
|
0.533
|
0.571
|
New York
|
12.725
|
9.393
|
9.520
|
10.018
|
10.227
|
9.445
|
9.318
|
10.792
|
9.477
|
North Carolina
|
1.896
|
3.261
|
2.766
|
2.823
|
2.619
|
2.954
|
2.891
|
2.817
|
2.828
|
North Dakota
|
0.800
|
0.273
|
0.246
|
0.256
|
0.302
|
0.215
|
0.254
|
0.251
|
0.275
|
Ohio
|
5.139
|
4.803
|
4.893
|
4.941
|
4.687
|
4.243
|
4.368
|
3.836
|
3.850
|
Oklahoma
|
0.791
|
1.275
|
1.236
|
1.224
|
1.152
|
1.207
|
1.219
|
1.186
|
1.207
|
Oregon
|
1.247
|
0.750
|
0.715
|
0.702
|
0.664
|
0.712
|
0.781
|
0.885
|
0.860
|
Pennsylvania
|
6.835
|
5.731
|
5.993
|
5.885
|
5.807
|
5.571
|
5.720
|
5.856
|
5.810
|
Rhode Island
|
0.691
|
0.665
|
0.635
|
0.615
|
0.670
|
0.753
|
0.712
|
0.857
|
0.803
|
South Carolina
|
0.683
|
1.349
|
1.278
|
1.260
|
1.201
|
1.394
|
1.403
|
1.288
|
1.323
|
South Dakota
|
0.649
|
0.235
|
0.249
|
0.253
|
0.272
|
0.233
|
0.240
|
0.242
|
0.257
|
Tennessee
|
1.386
|
1.801
|
1.743
|
1.717
|
1.700
|
1.865
|
1.848
|
1.730
|
1.771
|
Texas
|
2.264
|
6.524
|
7.668
|
7.349
|
7.135
|
7.183
|
6.942
|
6.529
|
6.870
|
Utah
|
0.748
|
0.599
|
0.559
|
0.508
|
0.413
|
0.452
|
0.494
|
0.568
|
0.526
|
Vermont
|
0.596
|
0.319
|
0.418
|
0.419
|
0.396
|
0.417
|
0.425
|
0.490
|
0.503
|
Virginia
|
1.957
|
3.041
|
2.428
|
2.486
|
2.490
|
2.581
|
2.607
|
2.588
|
2.692
|
Washington
|
2.051
|
1.204
|
1.225
|
1.245
|
1.145
|
1.244
|
1.305
|
1.443
|
1.434
|
West Virginia
|
0.906
|
0.907
|
0.663
|
0.639
|
0.638
|
0.625
|
0.631
|
0.661
|
0.707
|
Wisconsin
|
3.576
|
2.080
|
2.229
|
2.236
|
2.230
|
2.010
|
2.054
|
2.000
|
2.100
|
Wyoming
|
0.299
|
0.202
|
0.137
|
0.129
|
0.154
|
0.146
|
0.160
|
0.173
|
0.169
|
Source: State data were received by CRS from HHS in May 2007, September 2008, April 2009, June 2010, August 2011, August 2012, September 2013, February 2015, and November 2015.
Appendix A.
Estimated Allotments to the States Under Various Hypothetical Appropriations Levels
Table A-1, below, shows estimated allocations to the states at various hypothetical appropriations levels. In column (a) are allotments at the equivalent of a hypothetical FY1984 appropriation of $1.975 billion—under current LIHEAP practice where funds are set aside for leveraging incentive grants and training and technical assistance, the equivalent appropriation level is approximately $2.005 billion. The remaining columns show estimated allotments at appropriations of $2.14 billion, just under $2.25 billion, $2.25 billion, $3.0 billion, $3.39 billion, $4.0 billion, and $5.1 billion, the amount at which the LIHEAP program was last authorized in P.L. 109-58
.. In each
case, the estimates assume that 0.5% would be set aside for the territories, the amount set aside by
HHS starting in FY2014.
c11173008
Congressional Research Service
23
.
Table A-1. LIHEAP Estimated State Allotments for Regular Funds at Various Hypothetical Appropriation Levels
($ in millions)
“Old” Formula
Hypothetical
$1.975 Billion
in FY1984
(a)
State
“New” Formula, Hold-Harmless
Level Only
$2.14 Billion
(b)
Just Under
$2.25 Billion
(c)
“New” Formula, Hold-Harmless Level and Rate
$2.25 Billion
(d)
$2.5 Billion
(e)
$3.0 Billion
(f)
$3.39 Billion
(g)
$4.0 Billion
(h)
$5.1 Billion
(i)
Alabama
16.901
21.101
25.391
24.917
36.562
43.963
49.740
58.765
75.048
Alaska
10.788
10.788
10.841
11.351
12.629
15.185
17.181
20.298
25.923
Arizona
8.174
10.205
12.279
12.050
17.716
27.484
33.095
40.805
52.703
Arkansas
12.896
16.101
18.690
18.690
20.795
25.004
28.290
33.424
42.684
California
90.669
112.764
118.643
118.643
132.004
158.725
179.584
212.168
270.955
Colorado
31.613
31.613
31.613
31.613
34.198
41.120
46.524
54.965
70.195
Connecticut
41.241
51.490
59.889
59.889
66.634
80.122
90.651
107.099
136.774
Delaware
5.474
6.834
8.224
8.070
10.010
12.036
13.618
16.089
20.547
District of
Columbia
6.405
6.405
6.405
6.739
7.498
9.015
10.200
12.051
15.390
Florida
26.742
33.389
40.176
39.426
57.963
89.923
108.281
133.508
172.436
Georgia
21.144
26.399
31.765
31.173
45.828
71.098
85.613
105.558
136.337
Hawaii
2.129
2.659
3.199
3.139
4.615
6.484
7.336
8.667
11.068
Idaho
12.331
12.331
12.331
12.974
14.435
17.357
19.638
23.202
29.630
Illinois
114.147
114.147
114.147
114.147
114.147
120.431
136.258
160.980
205.584
Indiana
51.683
51.683
51.683
51.683
51.683
51.683
57.253
67.641
86.382
Iowa
36.628
36.628
36.628
36.628
36.628
36.628
36.628
39.682
50.677
Kansas
16.821
19.568
20.588
20.588
22.907
27.544
31.164
36.818
47.019
Kentucky
26.895
27.680
29.123
29.123
32.402
38.962
44.082
52.080
66.510
c11173008
CRS-24
.
“Old” Formula
Hypothetical
$1.975 Billion
in FY1984
(a)
State
“New” Formula, Hold-Harmless
Level Only
$2.14 Billion
(b)
Just Under
$2.25 Billion
(c)
“New” Formula, Hold-Harmless Level and Rate
$2.25 Billion
(d)
$2.5 Billion
(e)
$3.0 Billion
(f)
$3.39 Billion
(g)
$4.0 Billion
(h)
$5.1 Billion
(i)
Louisiana
17.279
21.573
25.958
25.474
30.379
36.528
41.328
48.827
62.356
Maine
26.717
26.717
26.717
26.717
26.717
31.088
35.173
41.555
53.069
Maryland
31.578
39.425
47.439
46.555
54.211
65.185
73.751
87.133
111.276
Massachusetts
82.495
92.276
97.087
97.087
108.020
129.886
146.956
173.619
221.725
108.373
108.373
108.373
108.373
111.453
134.014
151.625
179.136
228.771
Minnesota
78.076
78.076
78.076
78.076
78.076
78.076
78.076
78.076
92.151
Mississippi
14.490
17.315
18.218
18.218
20.270
24.373
27.576
32.579
41.606
Missouri
45.595
45.595
47.265
47.265
52.587
63.233
71.542
84.523
107.943
Montana
14.464
14.464
14.464
15.218
16.932
20.359
23.035
27.214
34.754
Nebraska
18.114
18.114
18.114
19.058
21.205
25.497
28.848
34.082
43.525
3.839
4.793
5.767
5.660
8.320
12.908
15.544
19.165
24.753
New
Hampshire
15.615
15.800
16.624
16.624
18.496
22.240
25.162
29.728
37.965
New Jersey
76.584
77.743
81.796
81.796
91.008
109.430
123.811
146.276
186.805
New Mexico
10.233
11.184
11.767
11.767
13.092
15.742
17.811
21.043
26.873
250.058
250.058
250.058
250.058
265.220
318.909
360.818
426.285
544.400
North
Carolina
37.266
46.528
55.986
54.942
69.232
83.246
94.186
111.276
142.108
North
Dakota
15.712
15.712
15.712
16.531
18.393
22.116
25.023
29.563
37.754
100.980
100.980
100.980
100.980
100.980
113.363
128.261
151.532
193.519
15.535
19.396
23.339
22.904
29.139
35.037
39.641
46.834
59.811
Michigan
Nevada
New York
Ohio
Oklahoma
c11173008
CRS-25
.
“Old” Formula
Hypothetical
$1.975 Billion
in FY1984
(a)
State
Oregon
“New” Formula, Hold-Harmless
Level Only
$2.14 Billion
(b)
Just Under
$2.25 Billion
(c)
“New” Formula, Hold-Harmless Level and Rate
$2.25 Billion
(d)
$2.5 Billion
(e)
$3.0 Billion
(f)
$3.39 Billion
(g)
$4.0 Billion
(h)
$5.1 Billion
(i)
24.502
24.502
24.502
24.502
24.502
26.162
29.600
34.970
44.660
Pennsylvania
134.318
134.318
134.318
134.318
143.931
173.067
195.810
231.338
295.437
Rhode Island
13.579
16.954
18.929
18.929
21.061
25.324
28.652
33.851
43.230
South
Carolina
13.423
16.759
20.165
19.789
29.093
38.049
43.050
50.861
64.953
South Dakota
12.761
12.761
12.761
13.426
14.938
17.962
20.323
24.010
30.663
Tennessee
27.245
34.016
38.217
38.217
42.521
51.128
57.847
68.343
87.279
Texas
44.490
55.548
66.839
65.593
96.430
149.602
180.144
222.112
286.876
Utah
14.691
14.691
14.691
15.457
17.197
20.679
23.396
27.641
35.300
Vermont
11.704
11.704
11.704
12.314
13.701
16.474
18.639
22.021
28.122
Virginia
38.465
48.025
57.155
56.709
63.592
76.465
86.513
102.210
130.531
Washington
40.302
40.302
40.302
40.302
40.302
42.629
48.231
56.982
72.771
West Virginia
17.799
17.799
17.799
18.727
20.836
25.053
28.346
33.489
42.768
Wisconsin
70.280
70.280
70.280
70.280
70.280
70.280
70.280
79.009
100.900
Wyoming
5.882
5.882
5.882
6.189
6.885
8.279
9.367
11.067
14.133
1,965.125
2,099.450
2,208.900
2,208.900
2,457.650
2,955.150
3,343.502
3,950.150
5,044.650
Total
Source: Congressional Research Service (CRS) calculations based on factors provided by the Department of Health and Human Services (HHS) in February 2015.
Notes: These estimates take into account current law, which allows HHS to set aside funds out of regular LIHEAP funds for territories, leverage incentive grants and
Residential Energy Assistance Challenge (REACH) grants and training and technical assistance. For each estimate, 0.5% is allocated to the territories, $27 million to
leveraging incentive and REACH grants, and $3 million to training and technical assistance. Differing allocations for these purposes could change state allotments.
c11173008
CRS-26
The LIHEAP Formula
.
Appendix B. Further Depiction of How State
Allotments Depend Upon Appropriation Levels
Figure B-1 graphically illustrates state allotments for three “typical” types of states over a range
of appropriations from $0 to $5.1 billion. Represented are (1) a hold-harmless level state, (2) a
hold-harmless level and rate state, and (3) a state whose increased allocations are ratably reduced
in order to maintain allocations for the hold-harmless level and rate states.
In the figure, there are three vertical areas. These areas separate the three levels of appropriations
that are triggers under current law and were explained previously in this report. The figure also
($ in millions)
"Old" Formula
|
"New" Formula, Hold-Harmless Level Only
|
"New" Formula, Hold-Harmless Level and Rate
|
State
|
Hypothetical $1.975 Billion in FY1984 (a)
$2.14 Billion (b)
Just Under $2.25 Billion(c)
$2.25 Billion (d)
$2.5 Billion (e)
$3.0 Billion (f)
$3.39 Billion(g)
$4.0 Billion (h)
$5.1 Billion (i)
Alabama
|
16.901
|
20.788
|
24.845
|
24.426
|
36.348
|
44.954
|
50.862
|
60.090
|
76.740
|
Alaska
|
10.788
|
10.788
|
10.788
|
11.351
|
12.629
|
15.185
|
17.181
|
20.298
|
25.923
|
Arizona
|
8.174
|
10.053
|
12.016
|
11.813
|
17.578
|
28.425
|
34.064
|
41.490
|
53.411
|
Arkansas
|
12.896
|
15.862
|
18.958
|
18.638
|
21.902
|
26.335
|
29.796
|
35.202
|
44.956
|
California
|
90.669
|
111.521
|
121.572
|
121.572
|
135.263
|
162.644
|
184.018
|
217.406
|
277.644
|
Colorado
|
31.613
|
31.613
|
31.613
|
31.613
|
34.739
|
41.771
|
47.260
|
55.835
|
71.306
|
Connecticut
|
41.241
|
50.725
|
55.392
|
55.392
|
61.629
|
74.105
|
83.843
|
99.056
|
126.502
|
Delaware
|
5.474
|
6.733
|
8.047
|
7.911
|
10.047
|
12.081
|
13.668
|
16.148
|
20.623
|
District of Columbia
|
6.405
|
6.405
|
6.405
|
6.739
|
7.498
|
9.015
|
10.200
|
12.051
|
15.390
|
Florida
|
26.742
|
32.893
|
39.313
|
38.649
|
57.513
|
93.003
|
111.453
|
135.747
|
174.751
|
Georgia
|
21.144
|
26.007
|
31.083
|
30.558
|
45.473
|
73.533
|
88.120
|
107.329
|
138.168
|
Hawaii
|
2.129
|
2.619
|
3.130
|
3.077
|
4.579
|
5.807
|
6.570
|
7.762
|
9.912
|
Idaho
|
12.331
|
12.331
|
12.331
|
12.974
|
14.435
|
17.357
|
19.638
|
23.202
|
29.630
|
Illinois
|
114.147
|
114.147
|
114.147
|
114.147
|
114.147
|
125.432
|
141.915
|
167.665
|
214.121
|
Indiana
|
51.683
|
51.683
|
51.683
|
51.683
|
51.683
|
52.952
|
59.911
|
70.781
|
90.393
|
Iowa
|
36.628
|
36.628
|
36.628
|
36.628
|
36.628
|
36.628
|
36.628
|
41.631
|
53.166
|
Kansas
|
16.821
|
20.625
|
21.701
|
21.701
|
24.145
|
29.032
|
32.847
|
38.807
|
49.560
|
Kentucky
|
26.895
|
29.277
|
30.804
|
30.804
|
34.273
|
41.210
|
46.626
|
55.086
|
70.349
|
Louisiana
|
17.279
|
21.252
|
25.401
|
24.972
|
34.262
|
41.197
|
46.611
|
55.068
|
70.326
|
Maine
|
26.717
|
26.717
|
26.717
|
26.717
|
26.717
|
31.496
|
35.635
|
42.101
|
53.766
|
Maryland
|
31.578
|
38.840
|
46.421
|
45.637
|
57.677
|
69.353
|
78.467
|
92.704
|
118.390
|
Massachusetts
|
82.495
|
94.497
|
99.424
|
99.424
|
110.620
|
133.013
|
150.492
|
177.798
|
227.062
|
Michigan
|
108.373
|
108.373
|
108.373
|
108.373
|
108.373
|
128.763
|
145.685
|
172.118
|
219.808
|
Minnesota
|
78.076
|
78.076
|
78.076
|
78.076
|
78.076
|
78.076
|
78.076
|
78.076
|
94.277
|
Mississippi
|
14.490
|
17.822
|
20.093
|
20.093
|
22.355
|
26.881
|
30.413
|
35.931
|
45.887
|
Missouri
|
45.595
|
45.595
|
47.390
|
47.390
|
52.727
|
63.401
|
71.733
|
84.748
|
108.230
|
Montana
|
14.464
|
14.464
|
14.464
|
15.218
|
16.932
|
20.359
|
23.035
|
27.214
|
34.754
|
Nebraska
|
18.114
|
18.114
|
18.114
|
19.058
|
21.205
|
25.497
|
28.848
|
34.082
|
43.525
|
Nevada
|
3.839
|
4.722
|
5.643
|
5.548
|
8.256
|
13.351
|
15.999
|
19.486
|
25.085
|
New Hampshire
|
15.615
|
16.536
|
17.398
|
17.398
|
19.357
|
23.276
|
26.335
|
31.113
|
39.734
|
New Jersey
|
76.584
|
79.065
|
83.186
|
83.186
|
92.554
|
111.290
|
125.915
|
148.761
|
189.980
|
New Mexico
|
10.233
|
11.989
|
12.614
|
12.614
|
14.035
|
16.876
|
19.094
|
22.558
|
28.809
|
New York
|
250.058
|
250.058
|
250.058
|
250.058
|
250.058
|
280.061
|
316.866
|
374.358
|
478.085
|
North Carolina
|
37.266
|
45.837
|
54.784
|
53.858
|
69.501
|
83.570
|
94.553
|
111.708
|
142.660
|
North Dakota
|
15.712
|
15.712
|
15.712
|
16.531
|
18.393
|
22.116
|
25.023
|
29.563
|
37.754
|
Ohio
|
100.980
|
100.980
|
100.980
|
100.980
|
100.980
|
113.773
|
128.725
|
152.081
|
194.219
|
Oklahoma
|
15.535
|
19.108
|
22.838
|
22.452
|
29.674
|
35.681
|
40.371
|
47.695
|
60.911
|
Oregon
|
24.502
|
24.502
|
24.502
|
24.502
|
24.502
|
25.412
|
28.752
|
33.968
|
43.380
|
Pennsylvania
|
134.318
|
134.318
|
134.318
|
134.318
|
142.795
|
171.701
|
194.265
|
229.513
|
293.106
|
Rhode Island
|
13.579
|
16.702
|
17.728
|
17.728
|
19.725
|
23.718
|
26.834
|
31.703
|
40.488
|
South Carolina
|
13.423
|
16.510
|
19.732
|
19.399
|
28.868
|
39.110
|
44.249
|
52.278
|
66.763
|
South Dakota
|
12.761
|
12.761
|
12.761
|
13.426
|
14.938
|
17.962
|
20.323
|
24.010
|
30.663
|
Tennessee
|
27.245
|
33.510
|
39.109
|
39.109
|
43.513
|
52.322
|
59.197
|
69.938
|
89.317
|
Texas
|
44.490
|
54.722
|
65.403
|
64.299
|
95.683
|
154.726
|
185.420
|
225.838
|
290.728
|
Utah
|
14.691
|
14.691
|
14.691
|
15.457
|
17.197
|
20.679
|
23.396
|
27.641
|
35.300
|
Vermont
|
11.704
|
11.704
|
11.704
|
12.314
|
13.701
|
16.474
|
18.639
|
22.021
|
28.122
|
Virginia
|
38.465
|
47.311
|
56.546
|
55.591
|
66.163
|
79.556
|
90.011
|
106.342
|
135.808
|
Washington
|
40.302
|
40.302
|
40.302
|
40.302
|
40.302
|
42.379
|
47.948
|
56.648
|
72.343
|
West Virginia
|
17.799
|
17.799
|
17.799
|
18.727
|
20.836
|
25.053
|
28.346
|
33.489
|
42.768
|
Wisconsin
|
70.280
|
70.280
|
70.280
|
70.280
|
70.280
|
70.280
|
70.280
|
82.945
|
105.927
|
Wyoming
|
5.882
|
5.882
|
5.882
|
6.189
|
6.885
|
8.279
|
9.367
|
11.067
|
14.133
|
Total
|
1,965.125
|
2,099.450
|
2,208.900
|
2,208.900
|
2,457.650
|
2,955.150
|
3,343.502
|
3,950.150
|
5,044.650
|
Source: Congressional Research Service (CRS) calculations based on factors provided by the Department of Health and Human Services (HHS) in November 2015.
Notes: These estimates take into account current law, which allows HHS to set aside funds out of regular LIHEAP funds for territories, leverage incentive grants and Residential Energy Assistance Challenge (REACH) grants and training and technical assistance. For each estimate, 0.5% is allocated to the territories, $27 million to leveraging incentive and REACH grants, and $3 million to training and technical assistance. Differing allocations for these purposes could change state allotments.
Appendix B.
Further Depiction of How State Allotments Depend Upon Appropriation Levels
Figure B-1 graphically illustrates state allotments for three "typical" types of states over a range of appropriations from $0 to $5.1 billion. Represented are (1) a hold-harmless level state, (2) a hold-harmless level and rate state, and (3) a state whose increased allocations are ratably reduced in order to maintain allocations for the hold-harmless level and rate states.
In the figure, there are three vertical areas. These areas separate the three levels of appropriations that are triggers under current law and were explained previously in this report. The figure also graphs the three basic types of states. Reading from top to bottom of Figure B-1, these three
types of states are as follows.
c11173008
•
Hold-Harmless Level Only States. These states are subject to only the
holdharmlesshold-harmless level provision. They do not qualify for the hold-harmless rate because
each state
’'s share of the regular funds at $2.25 billion is greater than 1%. An
example of a hold-harmless level only state is represented by the line that runs
from $0 to point G. The hold-harmless level is evident from point A to point F.
Here, despite increases in the appropriations level, the state allotment remains
fixed.
In In Table 3, these are the states that have a
“Y”"Y" in column (c) and an
“N”
"N" in column (e).
•
Ratable Reduction States. These states are subject to a ratable reduction. Their
new formula percentage is greater than their old (FY1984) percentage. An
example of these states is depicted by the line that runs from $0 to point H. There
is a small decrease in state allotments at point D that is attributable to the
increased shortfall on the distribution of funds that the hold-harmless
rate
imposes. In rate imposes. In Table 3, these are the states that have an
“N”"N" in both column (c) and
column (e).
•
Hold-Harmless Level
andand Rate States. These states are subject to both the
holdharmlesshold-harmless level and the hold harmless rate provisions. An example of a typical
level and rate state is shown by the line that runs from $0 to point I. The
holdharmlesshold-harmless level is evident by the fixed state allotment from point C to point E.
However, the (subtle) jump at exactly $2.25 billion signals that this state is
subject to the hold-harmless
raterate provision. After the allotment jump at $2.25
billion, the state
’'s allotment continues to increase (at a rate lower than the old
rate, but higher than the new rate). In Table 3
, , these are the states that have a
“Y” "Y" in column (c) and a
“Y”"Y" in the column (e).
Congressional Research Service
27
.
Figure B-1. Estimated LIHEAP Allocations at Various Hypothetical Appropriations Levels for Three Types of States
$120
Hold-Harmless Level
"Old"Formula
G
Hold-Harmless Rate
$100
Hold-Harmless
Level Only State
State Allotment
($ in millions)
$80
H
A
F
Ratably Reduced
State
$60
Hold-Harmless
Level and Rate
State
$40
I
D
$20
B
C
E
$0
$0
$1,000
$2,000
$3,000
Appropriation
($ in millions)
Source: Figure created by CRS using formula rates provided by HHS in
February 2015.
c11173008
CRS-28
$4,000
$5,000
The LIHEAP Formula
.
Appendix C. November 2015.
Appendix C.
LIHEAP Regular Fund Allocations to
the States, FY2009-FY2015, and Estimated FY2016
Allocations
Table C-1, Allocations
Table C-1, below, shows actual LIHEAP regular fund allocations to the states from FY2009
through FY2015 and estimated allocations for FY2016 based on the President
’'s budget request,
the FY2016 House Appropriations Committee-passed bill for the Departments of Labor, Health
and Human Services, and Education (LHE), and the Senate Appropriations Committee-passed
LHE bill.
For FY2016, the President’s budget proposed a total of $3.19 billion for LIHEAP regular funds.
Of that amount, $462 million would be distributed according to the “new” LIHEAP formula, with
the remainder, about $2.7 billion, distributed according to the “old” formula percentages. The
budget request would set aside $3 million for training and technical assistance and $27 million for
leveraging incentive and REACH grants. Column (h) of Table C-1 shows estimated allocations to
the states based on the President’s budget.
The House Appropriations Committee-passed bill would provide $3.365 billion for LIHEAP
regular funds, with $491 million distributed according to the “new” formula and nearly $2.9
billion using the “old” formula. The bill would also set aside not-quite $3 million for training and
technical assistance, and, while it does not specify an amount for leveraging incentive and
REACH grants, the bill authorizes funds to be used for this purpose in its reference to the
authorizing statute. Column (i) of Table C-1 shows estimated allocations to the states at the level
proposed in the House Committee-passed bill.
The Senate Appropriations Committee-passed bill would provide the same level of regular funds
appropriated in FY2015, $3.39 billion. Of the total, $491 million would be distributed using the
“new” LIHEAP formula and the remainder using the “old” formula. Like the House Committeepassed bill, the Senate Committee-passed bill authorizes funds to be used for leveraging incentive
and REACH grants, but does not specify an amount, and would provide nearly $3 million for
training and technical assistance. Column (j) of Table C-1 shows estimated allocations to the
states at the level proposed in the Senate Committee-passed bill.
LHE bill.
The FY2016 Consolidated Appropriations Act (P.L. 114-113) provided $3.39 billion in LIHEAP regular funds. The law provided that $2.988 million be set aside for training and technical assistance. As of the date of this report, HHS had not released total FY2016 funding allocations to the states, and it was unclear whether, as in recent years, funds would be set aside for leveraging incentive and REACH grants. As a result, column (h) of Table C-1 contains CRS estimates of allocations to the states for FY2016.
The FY2015 Consolidated and Further Continuing Appropriations Act (P.L. 113-235) provided
$3.39 billion in LIHEAP regular funds. Of the total, $491 million was distributed according to the
“new” "new" formula. Approximately $3 million was set aside for training and technical assistance, but
no funding was provided for leveraging incentive and REACH grants. Column (g) of Table C-1
contains allocations to the states in FY2015.
In FY2014, Congress appropriated approximately $3.425 billion for LIHEAP as part of the
Consolidated Appropriation Act (P.L. 113-76). Prior to distribution of funds, HHS reduced the
amount available by 1%, transferring $34.245
billionmillion within the agency. Of the $3.390 billion
available, HHS increased the amount available for the territories to 0.5% of the total; this was the
first time since the program
’'s inception that the territorial allocation changed from 0.134%. Of the
amount available to the states and tribes, $491 million was distributed according to the
“new”
"new" formula and the remainder according to the
“old”"old" formula. See column (f) of
Table C-1.
Table C-1.
c11173008
Congressional Research Service
29
The LIHEAP Formula
.
Column (e) of Table C-1
contains actual regular fund allocations to the states in FY2013. The
amount appropriated for LIHEAP as part of the FY2013 Consolidated and Further Continuing
Appropriations Act (P.L. 113-6) was the same as the FY2012 level—$3.472 billion for regular
funds, with $497 million distributed according to the
“new”"new" LIHEAP formula. However,
application of an across-the-board rescission of 0.2%, sequestration, and a transfer of funds
within HHS reduced the total amount available to $3.255 billion.
Column (d) contains actual allocations for FY2012 at an appropriations level of $3.472 billion
( (P.L. 112-74). The law provided a total of $3.478 billion for LIHEAP regular funds, but the
amount was reduced by an across-the-board rescission of 0.189% for discretionary accounts,
resulting in the $3.472 billion funding level. P.L. 112-74 also provided that, of the amount
appropriated, all but $497 million be distributed according to the proportions of the
“old”
"old" LIHEAP formula. In addition, $3 million was set aside for training and technical assistance.
In FY2009 (P.L. 110-329), FY2010 (P.L. 111-117), and FY2011 (P.L. 112-10) Congress
appropriated $4.51 billion for LIHEAP formula funds. Of this amount, $840 million was
distributed according to the
“new”"new" LIHEAP formula and the remaining funds, approximately
$3.67 billion, according to the
“old”"old" formula. Column (c) of Table C-1
shows the allocations to
the states in FY2011, column (b) shows allocations to the states in FY2010, and column (a)
shows FY2009 allocations. Note that funds were not distributed in exactly the same way in each
year for several reasons. LIHEAP formula rates are updated each year, which affects the
percentage of funds that states receive. In addition, two factors changed the FY2011
appropriation. The appropriations bill subjected all discretionary accounts to an across-the-board
rescission of 0.2%, and HHS did not distribute leveraging incentive and REACH grants, making
the total available to the states slightly more than in FY2009 and FY2010.
c11173008
Congressional Research Service
30
.
Table C-1. LIHEAP Actual State Regular Fund Allocations for
FY2009 through FY2015 and Estimated Allocations for FY2016
($ in millions)
Actual Allocations, FY2009-FY2015
FY2016
FY2016 House FY2016 Senate
President’s Appropriations Appropriations
Budget:
Committee:
Committee:
$3.19
$3.37
$3.39
Billionh
Billioni
Billionj
(h)
(i)
(j)
FY2009:
$4.51
Billiona
(a)
FY2010:
$4.51
Billionb
(b)
FY2011:
$4.50
Billionc
(c)
FY2012:
$3.47
Billiond
(d)
FY2013:
$3.26
Billione
(e)
FY2014:
$3.39
Billionf
(f)
FY2015:
$3.39
Billiong
(g)
Alabama
60.063
58.799
59.419
47.408
48.269
48.885
44.387
41.427
43.894
44.108
Alaska
23.568
25.308
23.667
18.002
17.171
18.841
17.482
16.411
17.357
17.494
Arizona
29.047
33.729
32.922
23.852
23.343
23.641
21.581
20.035
21.228
21.331
Arkansas
36.497
35.773
34.985
28.537
26.746
27.505
26.777
25.240
26.438
26.601
California
225.894
202.749
202.843
154.574
145.410
153.592
174.086
163.446
171.699
172.847
Colorado
63.474
64.257
62.139
47.308
44.270
46.378
48.889
45.318
48.057
48.457
Connecticut
95.783
96.942
98.254
79.532
76.014
77.413
85.764
80.847
84.678
85.200
Delaware
17.384
15.189
15.172
11.957
12.573
13.016
12.547
11.881
12.403
12.472
District of
Columbia
14.653
13.992
14.051
10.687
9.976
10.474
10.379
9.743
10.304
10.386
Florida
95.037
110.354
107.714
78.040
76.376
77.351
70.611
65.550
69.454
69.793
Georgia
75.141
87.252
85.164
61.702
60.387
61.158
55.829
51.827
54.914
55.182
Hawaii
4.652
6.023
6.027
6.107
5.416
6.159
5.622
5.219
5.530
5.557
Idaho
26.939
26.939
27.052
20.576
19.207
20.166
19.982
18.758
19.840
19.996
Illinois
237.236
232.865
238.712
185.684
160.191
167.458
167.396
155.951
164.390
165.835
Indiana
103.609
104.151
102.749
80.006
72.374
75.820
75.792
70.610
74.431
75.086
Iowa
67.803
67.803
68.137
54.813
51.292
53.735
53.715
50.043
52.750
53.214
Kansas
45.349
41.757
42.327
32.160
31.397
31.019
30.717
28.761
30.274
30.487
State
c11173008
FY2016 Estimated Allocations
CRS-31
.
FY2016 Estimated Allocations
Actual Allocations, FY2009-FY2015
FY2009:
$4.51
Billiona
(a)
FY2010:
$4.51
Billionb
(b)
FY2011:
$4.50
Billionc
(c)
FY2012:
$3.47
Billiond
(d)
FY2013:
$3.26
Billione
(e)
FY2014:
$3.39
Billionf
(f)
FY2015:
$3.39
Billiong
(g)
Kentucky
68.353
57.742
58.335
46.423
43.483
48.288
44.896
41.819
44.188
44.529
Louisiana
57.196
51.870
53.164
43.422
40.864
42.062
38.390
36.301
37.935
38.153
Maine
49.457
54.309
53.539
39.982
37.414
39.195
39.181
36.502
38.477
38.816
Maryland
101.296
82.002
85.523
69.790
70.390
68.513
68.854
65.052
68.023
68.422
Massachusetts
162.981
175.524
175.178
132.731
132.256
140.014
146.328
136.789
144.156
145.200
Michigan
222.412
233.524
228.294
173.450
165.582
165.444
161.827
149.653
158.974
160.345
Minnesota
144.528
144.528
145.241
116.839
109.335
114.541
114.498
106.670
112.443
113.431
Mississippi
39.011
39.661
38.834
31.591
29.313
30.120
26.996
25.306
26.615
26.798
Missouri
103.541
95.257
95.596
68.231
66.553
70.882
73.772
68.583
72.571
73.148
Montana
31.598
31.598
31.730
24.135
22.529
23.654
23.438
22.002
23.271
23.454
Nebraska
39.573
39.573
39.738
30.226
28.214
29.623
29.353
27.555
29.143
29.373
Nevada
13.643
15.841
15.462
11.203
10.964
11.104
10.136
9.410
9.970
10.019
New
Hampshire
34.112
34.112
34.255
26.055
24.321
25.536
25.750
23.967
25.339
25.536
New Jersey
166.690
177.196
180.991
136.746
124.480
124.570
126.586
117.839
124.570
125.539
New Mexico
24.901
22.355
22.448
17.074
15.938
16.734
17.844
16.665
17.575
17.704
New York
475.935
479.526
495.801
375.710
350.169
366.843
381.440
353.256
374.856
378.022
North
Carolina
123.243
109.339
111.263
83.011
87.702
88.271
86.504
81.955
85.523
85.995
34.325
34.325
34.469
26.218
24.473
25.695
25.460
23.901
25.279
25.478
220.588
223.108
225.398
165.463
144.794
154.314
148.087
137.962
145.428
146.706
State
North Dakota
Ohio
c11173008
FY2016
FY2016 House FY2016 Senate
President’s Appropriations Appropriations
Budget:
Committee:
Committee:
$3.19
$3.37
$3.39
Billionh
Billioni
Billionj
(h)
(i)
(j)
CRS-32
.
FY2016 Estimated Allocations
Actual Allocations, FY2009-FY2015
FY2016
FY2016 House FY2016 Senate
President’s Appropriations Appropriations
Budget:
Committee:
Committee:
$3.19
$3.37
$3.39
Billionh
Billioni
Billionj
(h)
(i)
(j)
FY2009:
$4.51
Billiona
(a)
FY2010:
$4.51
Billionb
(b)
FY2011:
$4.50
Billionc
(c)
FY2012:
$3.47
Billiond
(d)
FY2013:
$3.26
Billione
(e)
FY2014:
$3.39
Billionf
(f)
FY2015:
$3.39
Billiong
(g)
Oklahoma
49.007
47.902
47.717
36.094
35.955
37.147
36.338
34.439
35.929
36.126
Oregon
45.355
45.355
45.579
36.666
34.311
35.945
35.931
33.475
35.286
35.597
Pennsylvania
274.925
282.279
280.478
209.548
190.810
203.071
206.356
191.199
202.819
204.519
Rhode Island
30.209
29.666
29.790
23.241
23.976
23.813
27.361
25.753
27.004
27.176
South
Carolina
47.702
47.311
46.909
36.270
38.335
38.825
35.442
32.901
34.861
35.031
South Dakota
27.878
27.878
27.995
21.293
19.877
20.869
20.678
19.412
20.531
20.692
Tennessee
73.723
72.092
71.595
55.405
56.856
58.040
55.161
51.930
54.444
54.788
Texas
158.110
183.593
179.200
129.832
127.064
128.686
117.473
109.053
115.548
116.112
Utah
32.094
32.094
32.228
24.513
22.882
24.025
23.806
22.348
23.636
23.822
Vermont
25.568
25.568
25.675
19.529
18.230
19.140
18.965
17.804
18.830
18.978
118.084
100.856
102.839
80.436
78.971
81.877
81.432
76.829
80.420
80.906
Washington
74.603
74.603
74.971
60.310
56.437
59.124
59.102
55.062
58.041
58.551
West Virginia
40.584
38.884
39.047
29.700
27.723
29.108
28.842
27.076
28.636
28.861
Wisconsin
130.096
130.096
130.738
105.172
98.417
103.103
103.065
96.019
101.215
102.104
Wyoming
12.850
12.850
12.904
9.815
9.162
9.619
9.531
8.948
9.463
9.538
4,476.302
4,476.302
4,494.258
3,437.068
3,248.193
3,370.409
3,370.379
3,144.502
3,318.639
3,343.514
State
Virginia
Total
Source: The Department of Health and Human Services (HHS) provided data on final regular fund allocations for FY2008 through FY2015 (columns (a) through (g)).
Allocations to the states include tribal allotments, and FY2016 estimates assume that approximately 0.5% of the total would be set aside for the territories.
a.
c11173008
CRS-33
($ in millions)
Actual Allocations, FY2009-FY2015
|
FY2016 Estimated Allocations
|
State
|
FY2009: $4.51 Billiona (a)
FY2010:$4.51 Billionb(b)
FY2011:$4.50 Billionc(c)
FY2012: $3.47 Billiond(d)
FY2013:$3.26Billione(e)
FY2014:$3.39Billionf(f)
FY2015:$3.39Billiong(g)
FY2016 Appropriation: $3.39Billionh(h)
Alabama
|
60.063
|
58.799
|
59.419
|
47.408
|
48.269
|
48.885
|
44.387
|
44.250
|
Alaska
|
23.568
|
25.308
|
23.667
|
18.002
|
17.171
|
18.841
|
17.482
|
17.482
|
Arizona
|
29.047
|
33.729
|
32.922
|
23.852
|
23.343
|
23.641
|
21.581
|
21.400
|
Arkansas
|
36.497
|
35.773
|
34.985
|
28.537
|
26.746
|
27.505
|
26.777
|
27.882
|
California
|
225.894
|
202.749
|
202.843
|
154.574
|
145.410
|
153.592
|
174.086
|
177.340
|
Colorado
|
63.474
|
64.257
|
62.139
|
47.308
|
44.270
|
46.378
|
48.889
|
49.430
|
Connecticut
|
95.783
|
96.942
|
98.254
|
79.532
|
76.014
|
77.413
|
85.764
|
80.768
|
Delaware
|
17.384
|
15.189
|
15.172
|
11.957
|
12.573
|
13.016
|
12.547
|
12.584
|
District of Columbia
|
14.653
|
13.992
|
14.051
|
10.687
|
9.976
|
10.474
|
10.379
|
10.379
|
Florida
|
95.037
|
110.354
|
107.714
|
78.040
|
76.376
|
77.351
|
70.611
|
70.016
|
Georgia
|
75.141
|
87.252
|
85.164
|
61.702
|
60.387
|
61.158
|
55.829
|
55.358
|
Hawaii
|
4.652
|
6.023
|
6.027
|
6.107
|
5.416
|
6.159
|
5.622
|
5.575
|
Idaho
|
26.939
|
26.939
|
27.052
|
20.576
|
19.207
|
20.166
|
19.982
|
19.982
|
Illinois
|
237.236
|
232.865
|
238.712
|
185.684
|
160.191
|
167.458
|
167.396
|
167.396
|
Indiana
|
103.609
|
104.151
|
102.749
|
80.006
|
72.374
|
75.820
|
75.792
|
75.792
|
Iowa
|
67.803
|
67.803
|
68.137
|
54.813
|
51.292
|
53.735
|
53.715
|
53.715
|
Kansas
|
45.349
|
41.757
|
42.327
|
32.160
|
31.397
|
31.019
|
30.717
|
31.952
|
Kentucky
|
68.353
|
57.742
|
58.335
|
46.423
|
43.483
|
48.288
|
44.896
|
46.764
|
Louisiana
|
57.196
|
51.870
|
53.164
|
43.422
|
40.864
|
42.062
|
38.390
|
42.266
|
Maine
|
49.457
|
54.309
|
53.539
|
39.982
|
37.414
|
39.195
|
39.181
|
39.181
|
Maryland
|
101.296
|
82.002
|
85.523
|
69.790
|
70.390
|
68.513
|
68.854
|
72.314
|
Massachusetts
|
162.981
|
175.524
|
175.178
|
132.731
|
132.256
|
140.014
|
146.328
|
148.924
|
Michigan
|
222.412
|
233.524
|
228.294
|
173.450
|
165.582
|
165.444
|
161.827
|
158.928
|
Minnesota
|
144.528
|
144.528
|
145.241
|
116.839
|
109.335
|
114.541
|
114.498
|
114.498
|
Mississippi
|
39.011
|
39.661
|
38.834
|
31.591
|
29.313
|
30.120
|
26.996
|
29.079
|
Missouri
|
103.541
|
95.257
|
95.596
|
68.231
|
66.553
|
70.882
|
73.772
|
73.911
|
Montana
|
31.598
|
31.598
|
31.730
|
24.135
|
22.529
|
23.654
|
23.438
|
23.438
|
Nebraska
|
39.573
|
39.573
|
39.738
|
30.226
|
28.214
|
29.623
|
29.353
|
29.353
|
Nevada
|
13.643
|
15.841
|
15.462
|
11.203
|
10.964
|
11.104
|
10.136
|
10.051
|
New Hampshire
|
34.112
|
34.112
|
34.255
|
26.055
|
24.321
|
25.536
|
25.750
|
26.610
|
New Jersey
|
166.690
|
177.196
|
180.991
|
136.746
|
124.480
|
124.570
|
126.586
|
128.130
|
New Mexico
|
24.901
|
22.355
|
22.448
|
17.074
|
15.938
|
16.734
|
17.844
|
18.786
|
New York
|
475.935
|
479.526
|
495.801
|
375.710
|
350.169
|
366.843
|
381.440
|
366.707
|
North Carolina
|
123.243
|
109.339
|
111.263
|
83.011
|
87.702
|
88.271
|
86.504
|
86.773
|
North Dakota
|
34.325
|
34.325
|
34.469
|
26.218
|
24.473
|
25.695
|
25.460
|
25.460
|
Ohio
|
220.588
|
223.108
|
225.398
|
165.463
|
144.794
|
154.314
|
148.087
|
148.087
|
Oklahoma
|
49.007
|
47.902
|
47.717
|
36.094
|
35.955
|
37.147
|
36.338
|
36.874
|
Oregon
|
45.355
|
45.355
|
45.579
|
36.666
|
34.311
|
35.945
|
35.931
|
35.931
|
Pennsylvania
|
274.925
|
282.279
|
280.478
|
209.548
|
190.810
|
203.071
|
206.356
|
205.222
|
Rhode Island
|
30.209
|
29.666
|
29.790
|
23.241
|
23.976
|
23.813
|
27.361
|
26.027
|
South Carolina
|
47.702
|
47.311
|
46.909
|
36.270
|
38.335
|
38.825
|
35.442
|
35.143
|
South Dakota
|
27.878
|
27.878
|
27.995
|
21.293
|
19.877
|
20.869
|
20.678
|
20.678
|
Tennessee
|
73.723
|
72.092
|
71.595
|
55.405
|
56.856
|
58.040
|
55.161
|
56.152
|
Texas
|
158.110
|
183.593
|
179.200
|
129.832
|
127.064
|
128.686
|
117.473
|
116.483
|
Utah
|
32.094
|
32.094
|
32.228
|
24.513
|
22.882
|
24.025
|
23.806
|
23.806
|
Vermont
|
25.568
|
25.568
|
25.675
|
19.529
|
18.230
|
19.140
|
18.965
|
18.965
|
Virginia
|
118.084
|
100.856
|
102.839
|
80.436
|
78.971
|
81.877
|
81.432
|
83.999
|
Washington
|
74.603
|
74.603
|
74.971
|
60.310
|
56.437
|
59.124
|
59.102
|
59.102
|
West Virginia
|
40.584
|
38.884
|
39.047
|
29.700
|
27.723
|
29.108
|
28.842
|
28.842
|
Wisconsin
|
130.096
|
130.096
|
130.738
|
105.172
|
98.417
|
103.103
|
103.065
|
103.065
|
Wyoming
|
12.850
|
12.850
|
12.904
|
9.815
|
9.162
|
9.619
|
9.531
|
9.531
|
Total
|
4,476.302
|
4,476.302
|
4,494.258
|
3,437.068
|
3,248.193
|
3,370.409
|
3,370.379
|
3,370.379
|
Source: The Department of Health and Human Services (HHS) provided data on final regular fund allocations for FY2008 through FY2015 (columns (a) through (g)). Allocations to the states include tribal allotments, and FY2016 estimates assume that approximately 0.5% of the total would be set aside for the territories.
a.
Congress appropriated approximately $4.5 billion for LIHEAP as part of a continuing resolution (P.L. 110-329). Of this amount, $840 million was allocated under the
“new” "new" LIHEAP formula, with the remainder allocated according to the proportions of the
“old”"old" LIHEAP formula.
.
c11173008
b.
b.
In FY2010, Congress appropriated the same amount for LIHEAP regular funds as it had in FY2009—approximately $4.5 billion—with the same division of funds
between
“old” and “new”"old" and "new" formulas (P.L. 111-117). Although FY2010 LIHEAP funds were divided between the
“old” and “new”"old" and "new" formula in the same way as FY2009,
the awards to the states are different because the formula factors were updated in April 2009.
c.
c.
The FY2011 Department of Defense and Full-Year Continuing Appropriations Act (P.L. 112-10) included an across-the-board rescission of 0.2% for discretionary
accounts. This reduced the LIHEAP regular fund appropriation from approximately $4.51 billion to $4.50 billion. In addition, unlike appropriations in most years,
HHS did not set aside funds for leveraging incentive and REACH grants, and instead included these funds in the formula grants to the states, bringing the total
distributed to $4.49 billion.
d.
d.
The FY2012 Consolidated Appropriations Act (P.L. 112-74) included an across-the-board rescission of 0.189% that reduced the total available to $3,47 billion. Of
the amount appropriated, $497 million was distributed according to the
“new”"new" LIHEAP formula and the remainder according to the proportions of the
“old”
"old" LIHEAP formula. In addition, the law provided $3 million for training and technical assistance.
e.
e.
In FY2013, Congress enacted a full-year continuing resolution funding LIHEAP (and most other federal programs) at FY2012 levels (P.L. 113-6). While LIHEAP was
funded at $3.472 billion in FY2012, a series of deductions meant that the total available for LIHEAP in FY2013 was $3.255 billion.
f.
f.
The FY2014 regular fund appropriation for LIHEAP (P.L. 113-76) was reduced by 1% ($34.245 million) due to a transfer of funds within HHS, bringing the amount
available to $3.390 billion. HHS did not distribute leveraging incentive and REACH grants, and it increased the territorial allocation from 0.134% of total funds to
0.500%. Of the amount distributed to states and tribes by formula ($3.370 billion), $491 million was distributed according to the
“new”"new" formula and the remainder
according to the proportions of the
“old” formula.
g.
"old" formula.
g.
In FY2015, Congress appropriated $3.39 billion for LIHEAP regular funds as part of the Consolidated and Further Continuing Appropriations Act (P.L. 113-235). Of
the total, 0.5% was distributed to the territories, approximately $3 million went to training and technical assistance, and no funds were distributed for leveraging
incentive and REACH grants. Of the funds distributed to the states and tribes by formula, $491 million was distributed according to the
“new”"new" formula, and the
remainder, approximately $2.9 billion, according to the proportions of the
“old” formula.
h.
The President’s FY2016 budget proposed that $3.19 billion be appropriated for regular funds, with $467 million distributed via the “new” formula. The budget also
proposed that $3 million be set aside for training and technical assistance and $27 million for leveraging incentive and REACH grants.
i.
The House Appropriations Committee-passed bill would appropriate $3.365 billion for LIHEAP regular funds, with $491 million distributed using the “new” formula.
The appropriations bill does not specify an amount for leveraging incentive and REACH grants, but does give the authority for them to be distributed, so the
estimates assume that HHS would set aside $27 million for this purpose. The bill would provide $2.988 million for training and technical assistance.
j.
The Senate Appropriations Committee-passed bill would provide $3.39 billion for LIHEAP regular funds. Of that amount, $491 million would be distributed using
the “new” formula and the remainder using the “old” formula. The appropriations bill does not specify an amount for leveraging incentive and REACH grants, but
does give the authority for them to be distributed, so the estimates assume that HHS would set aside $27 million for this purpose. The bill would provide $2.988
million for training and technical assistance.
CRS-34
The LIHEAP Formula
.
Author Contact Information
Libby Perl
Specialist in Housing Policy
eperl@crs.loc.gov, 7-7806
c11173008
Congressional Research Service
35
"old" formula.
h.
As of the date of this report, HHS had not released final allocations based on the FY2016 Consolidated Appropriations Act (P.L. 114-113). Figures in column (h) are CRS estimates. The estimates assume that, as in recent years, no funding will be allocated to leveraging incentive and REACH grants, and that $2.988 million will be set aside for training and technical assistance.
Author Contact Information
[author name scrubbed], Specialist in Housing Policy
([email address scrubbed], [phone number scrubbed])
Footnotes
1.
|
For additional information on LIHEAP, see, LIHEAP: Program and Funding, by [author name scrubbed].
|
2.
|
LIHEAP is codified at 42 U.S.C. §§8621-8630.
|
3.
|
U.S. Department of Health and Human Services, Administration for Children and Families, FY2010 LIHEAP Report to Congress, p. 35.
|
4.
|
The formula section is codified at 42 U.S.C. §8623.
|
5.
|
Depending on how Congress appropriates them, contingency funds may remain available for distribution in more than one fiscal year or they may expire with the fiscal year for which they were appropriated.
|
6.
|
The statutory definition of emergency includes a significant home energy supply shortage or disruption, a significant increase in the cost of home energy, a significant increase in home energy disconnections, a significant increase in participation in a public benefit program, a significant increase in unemployment, or an event meeting such criteria as the Secretary determines to be appropriate. 42 U.S.C. §8622.
|
7.
|
Community Services Administration, "Character and Scope of Specific Community Action Programs: Emergency Energy Conservation Program," Federal Register, vol. 40, no. 145, July 28, 1975, p. 31603.
|
8.
|
See, for example, House Appropriations Committee, report to accompany H.R. 4877, the FY1977 Supplemental Appropriations Act, 95th Cong., 1st sess., H.Rept. 95-68, March 11, 1977: "The funds in this program are used primarily to purchase materials to insulate the homes of low-income families."
|
9.
|
For example, if a southern state experiences 700 heating degree days in a year and a northern state experiences 7,000, the northern state has 10 times as many heating degree days as the southern state. However, if both numbers are squared, the northern state has 100 times as many heating degree days as the southern state.
|
10.
|
Community Services Administration, "Emergency Energy Conservation Program: Submission of Funding Plans," Federal Register, vol. 41, no. 208, October 27, 1976, p. 47096.
|
11.
|
Ibid.
|
12.
|
Ibid., pp. 47096-47097.
|
13.
|
Community Services Administration, "Special Crisis Intervention Program: General Information, Application Procedures, and Post Grant Requirements," Federal Register, vol. 42, no. 125, June 29, 1977, p. 33240.
|
14.
|
The formula was described in the Senate Appropriations Committee report to accompany H.R. 4877, the FY1977 Supplemental Appropriations Act, 95th Cong., 1st sess., S.Rept. 95-64, March 24, 1977. The CSA implemented this formula, which it described in guidance to the states. See the Federal Register, Ibid.
|
15.
|
Funds were appropriated through the FY1978 Supplemental Appropriations Act (P.L. 95-240) and in FY1979 through a continuing resolution (P.L. 95-482). In FY1978, Congress called the program Emergency Energy Assistance Program and in FY1979 called it the Crisis Intervention Program (excluding the word "Special" from the title).
|
16.
|
Community Services Administration, "Emergency Energy Conservation Program: Funding Requirements for Emergency Energy Assistance Program," Federal Register, vol. 43, no. 46, March 8, 1978, p. 9476.
|
17.
|
Community Services Administration, "Emergency Energy Conservation Program: Fiscal Year 1979 Crisis Intervention Program," Federal Register, vol. 43, no. 250, December 28, 1978, pp. 60466-60467.
|
18.
|
Congress appropriated $250 million for ECAP as part of an FY1980 Continuing Resolution (P.L. 96-123, referencing the FY1980 Departments of Labor, Health and Human Services and Education Appropriations bill, H.R. 4389), and appropriated an additional $150 million as part of the Department of the Interior Appropriations Act (P.L. 96-126).
|
19.
|
"Windfall Profits Tax." In CQ Almanac 1979, 35th ed., 609-32 (Washington, DC: Congressional Quarterly, 1980) http://library.cqpress.com/cqalmanac/cqal79-1184031.
|
20.
|
Senate Committee on Labor and Human Resources, Home Energy Assistance Act, report to accompany S. 1724, 96th Cong., 1st sess., S.Rept. 96-378, October 25, 1979, p. 12.
|
21.
|
See, for example, Senator Russell Long, "Home Energy Assistance Act," Senate debate, Congressional Record, vol. 125, part 25 (November 14, 1979), p. 32278. "But the formula [as passed by the Senate Finance Committee] went a long way toward considering the total household expense for energy, not just heating."
|
22.
|
Senator Rudy Boschwitz, "Home Energy Assistance Act," Senate debate, Congressional Record, vol. 125, part 25 (November 14, 1979), p. 32290. "I refer back to the committee report, which talks about the intent of the act being to 'offset high heating costs (and cooling where medically necessary) and that assistance not be a supplement of all utilities and their use to run appliances, etc.'... It is very clear that it is the intent of the Senate to help keep people warm."
|
23.
|
Senator Edmund Muskie, "Home Energy Assistance Act," Senate debate, Congressional Record, vol. 125, part 25 (November 14, 1979), p. 32288. "I do not often do this. As a matter of fact, this is my 21st year in the Senate, and I can recall only one other time in which I have sought to use delay and extended debate to make a point and to achieve justice. I am not a filibusterer. If I did not believe deeply about this, I would not be standing here."
|
24.
|
The BLS determined the lower living standard income level through its annual family budgets, which it maintained from 1947 to 1981. At the time the LIEAP program was enacted, the BLS developed annual family budgets assuming three different standards of living: lower, intermediate, and higher. The budget was calculated using costs of consumer goods including food, housing, transportation, clothing, and health care (unlike the federal poverty guidelines, which are based on the amount of money needed to buy food). The budget was then adjusted for family size and the prices of goods in various cities throughout the country. See David S. Johnson, John M. Rogers, and Lucilla Tan, "A Century of Family Budgets in the United States," Monthly Labor Review, 124, no. 5 (May 2001): 28-45.
|
25.
|
Senator Russell B. Long, "Home Energy Assistance Act," Senate debate, Congressional Record, vol. 125, part 25 (November 15, 1979), p. 32561. "This language was evolved in the Finance Committee. When the majority of the committee voted to exclude such items as air-conditioning and anything related to cooling a house and limited that formula to heating, this Senator contended that, if that were to be the case, there should be at least a minimum on which people could depend."
|
26.
|
According to the law, "The State is authorized to make grants to eligible households to meet the rising costs of cooling whenever the household establishes that such cooling is the result of medical need pursuant to standards established by the Secretary."
|
27.
|
House Committee on Appropriations, report to accompany H.R. 7998, the FY1981 Departments of Labor, Health and Human Services, and Education Appropriations Act, 96th Cong., 2nd sess., H. Rept. 96-1244, August 21, 1980, pp. 75-76.
|
28.
|
See, for example, Representative David Obey, "Low Income Energy Assistance," House debate, Congressional Record, vol. 126, part 18 (August 27, 1980), p. 23505. "Last year the Congress adopted a formula which, very frankly, was unfair to the South. It provided a much larger amount of the money available than it should have to Northern States. In response to that, Senator Long, on the windfall profit tax legislation, adopted an amendment which, for the block grant portion of the program, provided phenomenal increases for the Southern States at the expense of the Northern States."
|
29.
|
"The Low-Income Home Energy Assistance Program: An Analysis of the 1984 Reauthorization Issues," Coalition of Northeastern Governors, April 1984, p. 5.
|
30.
|
H.Rept. 96-1244 did not specify the years between which the increase in home heating expenditures should be measured. In implementing the formula, HHS measured the increase between 1978 and 1980.
|
31.
|
See, for example, Senate debate, Congressional Record, vol. 125, parts 24-25 (November 13-15, 1979), pp. 32082-32086, 32275-32293, 32558-32565, and 32576-32589.
|
32.
|
House debate, Congressional Record, vol. 126, part 18 (August 27, 1980), pp. 23502-23515.
|
33.
|
See, for example, Comments of Rep. Billy Tauzin, U.S. Congress, Joint Hearing before the Subcommittees of the Committees on Energy and Commerce, Education and Labor, and Ways and Means, Energy Costs and Low Income Energy Assistance, 98th Cong., 1st sess., February 24, 1983, pp. 119-120.
|
34.
|
Report of the Committee on Energy and Commerce to accompany H.R. 2439, the Low-Income Home Energy Assistance Amendments of 1984, 98th Cong., 2nd sess., H.Rept. 98-139, Part 2, May 15, 1984, p. 13.
|
35.
|
Ibid., p. 4.
|
36.
|
Congressional Record, vol. 129, part 17 (September 13, 1983), p. 23877. The greatest increases in percentage allocations were for Florida at 51%, Texas at 44%, and Alabama at 37%. The states whose percentage allocations decreased the most were Vermont at 32%, North Dakota at 24%, and New Hampshire at 23%.
|
37.
|
"The Low-Income Home Energy Assistance Program: An Analysis of the 1984 Reauthorization Issues," Coalition of Northeastern Governors, April 1984, p. 9.
|
38.
|
The final version of S. 2565 can be found in the Congressional Record, daily edition, vol. 130 (October 4, 1984), p. S13393.
|
39.
|
Congressional Record, daily edition, vol. 130 (October 4, 1984), pp. S13415-S13416.
|
40.
|
Report of the Committee on Energy and Commerce to accompany H.R. 2439, the Low-Income Home Energy Amendments of 1984, 98th Cong., 2nd sess., H.Rept. 98-139, Part 2, May 15, 1984, p. 13.
|
41.
|
"[A] State's allotment percentage is the percentage which expenditures for home energy by low-income households in that State bears to such expenditures in all States." 42 U.S.C. §8623(a)(2).
|
42.
|
The LIHEAP statute considers households with income at or below 150% of poverty or 60% of state median income (whichever value is greater) to be low income. 42 U.S.C. §8624(b)(2)(B).
|
43.
|
A state's heating and cooling degree data are weighted by population in the state.
|
44.
|
For more information about the RECS, see the EIA website at http://www.eia.doe.gov/emeu/recs/.
45.
|
The EIA's state data tables are available at http://www.eia.doe.gov/emeu/states/_seds.html.
|
46.
|
When appropriations are below a hypothetical FY1984 appropriation of $1.975 billion, the result of the current law's hold-harmless provisions is that states receive the same allotment percentages that they did under the old formula. See U.S. Department of Health and Human Services, Low Income Home Energy Assistance Program: Report to Congress for FY1987, p. 133.
|
47.
|
This amount is arrived at by adding $27 million and $3 million to $1.975 billion.
|
48.
|
HHS Administration for Children and Families, Office of Community Services, LIHEAP Dear Colleague Notice Allocation for Territories FY2014, November 22, 2013, http://www.acf.hhs.gov/programs/ocs/resource/liheap-allocation-for-territories-fy-2014.
|
49.
|
The exceptions to this are states that benefit from the hold-harmless rate, described in the next section, and the states that are ratably reduced in order to compensate states that benefit from the hold-harmless rate.
|
50.
|
"States which thereby receive the greatest proportional increase in allotments ... shall have their allotments reduced to the extent necessary to ensure that ... no State for fiscal year 1986 and thereafter shall receive less than the amount of funds the State would have received in fiscal year 1984." 42 U.S.C. §8623(a)(2)(A)(ii).
|
51.
|
"[A]ny State whose allotment percentage out of funds available to States from a total appropriation of $2,250,000,000 would be less than 1 percent, shall not, in any year when total appropriations equal or exceed $2,250,000,000, have its allotment percentage reduced from the percentage it would receive from a total appropriation of $2,140,000,000." 42 U.S.C. §8623(a)(2)(B).
|
52.
|
For more information about this issue, see Appendix C of this report.