Colorado’s Medicaid expansion, which began in January 2014, was larger – and costlier – than almost anyone anticipated
Expansion costs added up to nearly $1.6 billion during the first two years — 29 percent more than the $1.2 billion forecast. Nearly all of these costs were covered by the federal government.
Costs exceeded projections primarily due to the unexpectedly high enrollment, with caseload growth 71 percent higher than anticipated.
The moment Colorado decided to expand Medicaid under the Affordable Care Act (ACA) in 2013, experts got to work forecasting what the state should expect. But none of the predictions fully captured what happened.
Colorado’s Medicaid expansion, which began in January 2014, was larger – and costlier – than almost anyone anticipated. And attention is now turning to the state’s long-term costs as Colorado prepares to begin picking up some of the Medicaid expansion tab from the federal government.
So far, the federal government has been covering all of the costs for newly eligible Medicaid clients. But it will gradually reduce its contribution beginning next year. By 2020, the ongoing federal payment rate will be 90 percent.
The Colorado Health Institute (CHI) estimates that Colorado will be spending $222 million annually for Medicaid expansion by 2020 — its 10 percent share for the newly eligible Medicaid clients as well as its 50 percent share for “welcome mat” clients, those who were already eligible but didn’t enroll until the expansion began.
It is anticipated, however, that a majority of those state dollars will come from the Hospital Provider Fee rather than the General Fund. (See Figure 6.)
Additional findings based on CHI’s analysis of publicly available data from the state, include:
- The Medicaid expansion population stood at 289,000 by June 2015 and will reach an estimated 363,000 by June 2017.
- Expansion costs added up to nearly $1.6 billion during the first two years — 29 percent more than the $1.2 billion forecast. Nearly all of these costs were covered by the federal government.
- Costs exceeded projections primarily due to the unexpectedly high enrollment, with caseload growth 71 percent higher than anticipated.
- Per capita costs, on the other hand, were lower than predicted. On average, each expansion enrollee cost approximately $4,100 annually in the first two years compared with the anticipated annual cost of $5,200.
- Children have had the lowest per capita costs, averaging about $1,500 per year.
Newly eligible low-income adults without dependent children accounted for 77 percent of the expansion enrollees but 86 percent of the expansion spending. They averaged about $4,600 in annual costs.
But costs are only half the equation. It is important to remember what Colorado is getting in return. Medicaid coverage will increase the health and economic well
being of enrollees, with many clients gaining health insurance for the first time.
Some analyses even indicate that the state’s investment could end up saving money in the long run. Reduced spending on programs for the uninsured, increased employment in the health care sector and economic growth have all been cited as potential benefits in a post-expansion world.
This brief answers key questions based on CHI’s analysis of the expansion enrollment and spending so far: What is Medicaid expansion costing Colorado now and what will it cost in the future? And what is the state getting in return?
Sudden Growth: The First Two Years
Colorado saw 289,000 residents enroll in Medicaid due to expanded eligibility in FY 2013-14 and FY 2014-15.2
This expansion group made up nearly one of four (23 percent) of the state’s 1.2 million Medicaid beneficiaries by June 2015.2 Most of the expansion enrollees were newly eligible for the program. But 12 percent, or one of eight, were “welcome mat” enrollees, which means they were already eligible under pre-expansion criteria but hadn’t yet signed up for Medicaid. (See Figure 1.)
The difference is crucial because the federal government reimburses half of the cost for the care of welcome mat enrollees, the same rate as the rest of the state’s traditional Medicaid enrollees, not the full 100 percent that is reimbursed for the newly eligible.
The expansion costs added up to nearly $1.6 billion during the first two years — 29 percent more than the $1.2 billion forecasted by the fiscal note for the authorizing legislation.
Costs exceeded projections primarily due to the unexpectedly high enrollment. Caseload growth was 71 percent higher than anticipated. But per capita costs were actually lower than predicted, which kept total costs from exceeding forecasts even more.
On average, each expansion enrollee — both newly eligible and welcome mat enrollees — cost approximately $4,100 annually in the first two years. Projections had anticipated spending approximately $5,200 per enrollee.
Per capita expenses have varied greatly across different Medicaid populations. Children have been the biggest bang for the state buck under expansion, costing $1,500 per child per year. On the other end of the spectrum, former foster children — foster children who were enrolled in Medicaid and may now continue their enrollment to age 26 — have cost upward of $6,500 per capita.
The biggest miss by experts was the overall cost and enrollment for newly eligible low-income adults without dependent children (AwDC). Spending on this group, with family incomes at or below 138 percent of the federal poverty level (FPL), or about $16,500 a year, exceeded predictions by $209 million between January 1, 2014 and June 30, 2015.
This population of 222,000 Coloradans accounted for 77 percent of the expansion enrollees but 86 percent of the expansion spending, or $1.3 billion. (See Figure 2.)
Adults without kids were one of the more expensive groups on a per capita basis, costing an average of $4,600 a head. (See Figure 3.)
Why the large price tag for adults without dependent children? As one of the last groups to become eligible for Medicaid, they were the most likely to be in severe poverty and unable to afford care before expansion.
For example, parents with incomes up to 60 percent of the FPL were eligible for Medicaid coverage prior to expansion. But childless adults at or below that same poverty level could not enroll. They are now new to the Medicaid program and are more likely to carry the complex — and costly — health issues often associated with people who put off care because they couldn’t afford it.
The State Fiscal Years
This paper frequently refers to Colorado’s fiscal year. The state fiscal year runs from July to June. For example, fiscal year 2013-14 is defined as July 1, 2013 through June 30, 2014.
Terms used to discuss Medicaid expansion can be opaque and inconsistent. These are key definitions used throughout this analysis.
• Expansion Population: Those who enrolled as a result of the expansion of Medicaid eligibility outlined in the ACA and authorized by the Colorado General Assembly in 2013. This includes both the newly eligible and the welcome mat populations. In order to align with estimates from the Colorado Department of Health Care Policy and Financing (HCPF), this brief does not assess eligibility expansion under the Colorado Health Care Affordability Act of 2009 (HB09-1293) that was in effect prior to January 1, 2014.
• Newly Eligible: Enrollees who qualified after the ACA expanded eligibility to more low-income people, including adults without dependent children with incomes at or below 138 percent of the FPL and parents earning between 60 percent and 138 percent of the FPL.
• Welcome Mat: Medicaid enrollees who were already eligible under pre-expansion criteria but hadn’t enrolled in the program. Many signed up because of the extra attention, media coverage and outreach resulting from the ACA. Many welcome mat enrollees were children with newly eligible parents.
• Traditional Medicaid: Medicaid enrollees who were already eligible under pre-expansion criteria and whose enrollment was most likely unaffected by the ACA.
• Hospital Provider Fee: A fee paid by hospitals to the state that generates federal matching funds. These dollars are pooled in the Hospital Provider Fee fund, which goes toward Medicaid expansion and administration as well as hospital reimbursements.4
Sharing the Cost: The Role of State Funding
The breakdown in Medicaid funding is complex and varies by client group.
Colorado lawmakers voted to expand Medicaid eligibility in 2009, ahead of the ACA passage, extending coverage to parents below the FPL and childless adults under 10 percent of the FPL. HB09-1293, which authorized the state expansion, also established the Hospital Provider Fee fund to pay for it.
Because Colorado’s expansion began before the ACA- authorized financing kicked in, the state received its standard Medicaid match of 50 percent from the federal government and paid for the other half from its Hospital Provider Fee revenue.
In January 2014, the federal government began to pay nearly all of the costs for people brought into the program under the state expansion as well as those who became newly eligible under the ACA criteria.
Most of the ACA Medicaid expansion expenses did not come from state coffers. Instead, the cost was primarily covered by the federal government because of how the Medicaid eligibility expansion is funded under the ACA.
Since only half of FY 2013-14 occurred during expansion, the vast majority of the $1.6 billion cost of Medicaid expansion in the first two years was spent during FY 2014-15, the first full fiscal year of expansion.
The state received $1.1 billion from the federal government to cover the newly eligible population over the course of those 12 months.
Another $40 million came from the federal government to cover its half share for welcome mat enrollees.
Colorado’s General Fund was responsible for the other $40 million for the welcome mat population.
Going forward, the state will continue to be responsible for 50 percent of the welcome mat population as well as 50 percent of the traditional Medicaid population.
These expenditures will continue to come from the state General Fund. (See Table 1)
But the state will cover five percent of the cost of the newly eligible expansion group in calendar year 2017. States’ shares are set to increase to six percent in 2018, seven percent in 2019 and 10 percent in 2020. Colorado intends to tap money from the Hospital Provider Fee (shown in light blue) to pay for these costs. Because Colorado’s fiscal years do not align with calendar years, this will effectively mean a state share of 2.5 percent in FY 2016-17, 5.5 percent in FY 2017-18, and so on (see Figure 6).
The federal government (shown in dark blue) will still cover the majority of Medicaid-related expenditures, even as its share for expansion ramps down from 95 percent in 2017 to 90 percent in 2020 and after.
New Expectations: The Next Two Years
Medicaid enrollment among the expansion populations is expected to continue to grow, but at a slower rate. An additional 74,000 Coloradans will enroll in Medicaid under the expansion criteria by June 30, 2017, according to CHI projections.
Combined with the 289,000 who enrolled in the first two years, this will bring the total number of expansion enrollees to 363,000.
The federal government will continue to pay around $1.5 billion a year for Colorado’s Medicaid expansion in FY 2015-16 and FY 2016-17, according to a CHI analysis of projections from the Department of Health Care Policy and Financing (HCPF).
In five years, when the state’s share of newly eligible population spending is at the full 10 percent, CHI estimates that Colorado will pay $168 million for the newly eligible population plus $54 million for welcome mat enrollees.
That totals $222 million in state funds in FY 2020-21 to pay for Medicaid expansion in Colorado.
CHI expects state costs will be at least this high in subsequent years.
Medicaid Expansion: The Bottom Line
Clearly, Medicaid expansion in Colorado has cost much more than initially projected.
And even with the federal government footing most of the bill, state policymakers will continue to confront tough budget choices as Colorado’s portion of the cost increases over the next years.
The budget for HCPF, the state department that administers Medicaid, has accounted for a larger portion of the state’s General Fund over the past decade, increasing from 17 percent in FY 1999-2000 to 24 percent in the FY 2016-17 budget request. (See Figure 7.) Other parts of the budget, including education, have shrunk over that time.
A good deal of this increase for HCPF relates to the cost of Medicaid. Not only will Colorado’s expansion require additional spending, the state’s traditional,
non-expansion Medicaid spending — which currently averages a little over $4 billion annually7 — will continue to grow as well.
While it is important to analyze the costs of Medicaid expansion, it is also important to consider the benefits.
By mid-2017, 363,000 Coloradans will have gained health insurance through Medicaid due to expansion. And that coverage can have important implications.
Research by the National Academies’ Health and Medicine Division (formerly the Institute of Medicine) has found that access to care, quality of care, sickness and even life expectancies are worse for uninsured Americans.
In addition to health benefits, Medicaid expansion will likely improve the economic well-being of many state residents. For example, debt due to medical bills is the most common reason Americans declare bankruptcy.9 And while safety net clinics and hospitals provide care for many underserved groups, research by the Urban Institute has found that this system “does not fully substitute for health insurance.”
States may save money by reducing spending on programs for the uninsured. Some research has found that these cost savings outweigh the price of Medicaid expansion, even as states begin to pay a larger share.
Large uninsured populations can also be a burden to hospitals. One report by the American Hospital Association found that in 2010 community hospitals spent nearly $40 billion on uncompensated care — about six percent of their total expenses.
Hospitals don’t bear these costs alone. They may be passed on indirectly to those with private insurance in the form of increased premiums.
Finally, a recent report commissioned by the Colorado Health Foundation suggests that Medicaid expansion has greatly benefited the state economy, increased employment and pushed up the average household income. The analysis found that “increased tax revenue due to the larger post-expansion economy and modest savings in other state programs has and will allow Colorado to support this expansion at no cost to the state’s General Fund.”
Colorado isn’t the only state to anticipate these positive effects. Business leaders and chambers of commerce from many different states have lobbied their legislators to opt into Medicaid expansion. They feel this additional Medicaid funding will help support jobs, and increased coverage will create a healthier workforce. Many businesses currently bear the cost for the uninsured by paying higher insurance premiums for their employees.
Benefits to enrollees, economic stimulations and cost savings were part of Governor John Hickenlooper’s rationale when he first proposed implementing the ACA expansion.
The decision to expand Medicaid was more costly and more beneficial than almost anyone expected. Effects on the state’s economy and residents will continue to play out for many years to come.
Medicaid: A Primer
Medicaid is a joint federal and state program that covers the health care costs of low-income people. In Colorado, the state government and the federal government each pay about half, although the proportions are different in other states.
The ACA provided financial incentives for states to expand Medicaid and fill in coverage gaps that had left some low-income populations without insurance. The federal government promised to pay all costs of covering the newly insured in 2014, 2015 and 2016. The state’s share will start at five percent in 2017, growing to 10 percent by 2020 where it will remain.
Because of a U.S. Supreme Court ruling, the expansion was optional for states. So far, 31 states plus the District of Columbia have opted to expand the program.5
Here’s a quick look at the main eligibility guidelines under expansion:
|Pre-Expansion||Colorado HB09-1293 Expansion||ACA Expansion|
|Children||0-147% FPL||0-147% FPL||0-147% FPL|
|Adults Without Dependent Children||Not qualified||0-10% FPL||0-138% FPL|
|Parents||0-60% FPL||0-100% FPL||0-138% FPL|
|Pregnant Women||0-200% FPL||0-200% FPL||0-200% FPL|
The Colorado Health Institute (CHI) based this analysis on data published by the Colorado Department of Health Care Policy and Financing (HCPF) in December 2015. CHI made additional projections or assumptions about these data as necessary for this analysis. The assumptions are detailed here.
HCPF has published projections of expansion population costs and enrollment that distinguish between the welcome mat and newly eligible populations only through FY 2015-16. CHI projected the costs from FY 2016-17 through FY 2020-21 based on two assumptions: that spending for welcome mat enrollees will grow at an annual rate of 3.8 percent, the growth rate predicted for the non-expansion population between FY 2015-16 and FY 2016-17; and that spending on newly eligible enrollees will grow at an annual rate of 3.5 percent, the spending growth rate that HCPF projects for this population between FY 2015-16 and FY 2016-17.
CHI projected additional Medicaid enrollment through FY 2016-17 based on two assumptions: that enrollment for the welcome mat population will grow at an annual rate of 5.1 percent, the same projected growth rate of the non-expansion population between FY 2015-16 and FY 2016-17; and that enrollment for newly eligible enrollees will grow at an annual rate of 3.8 percent, the same growth rate projected for this population between FY 2015-16 and FY 2016-17.
CHI used HCPF data from December 2015 in this analysis. In a later analysis, from February 2016, HCPF revised upward its estimated Medicaid caseload.
However, for consistency, the data used by CHI in this analysis are based on HCPF’s responses to the Joint Budget Committee on December 16, 2015.
In a few instances, CHI made additional assumptions regarding federal-state match rates related to Medicaid expansion payments.
This analysis assumes a 100 percent federal match rate for all newly eligible enrollees. In reality, HCPF estimates that a tiny portion of the newly eligible population— 0.5 percent — hovered near the pre-expansion eligibility level and most likely would have qualified for Medicaid with the extra step of undergoing an “assets test.”The federal government pays 87.7 percent of the cost of this population rather than the 100 percent for other newly eligible enrollees.
CHI assumes a 50 percent federal match for welcome mat enrollees, which has been the case in Colorado for some time. The federal match, however, is subject to change because it is based on the ratio of state per capita income to the average national per capita income. Kaiser Family Foundation estimates Colorado’s match is now 50.7 percent and that it will vary between 50 and 51 percent in the next few years.18
Colorado’s Hospital Provider Fee fund covered the cost of the state’s match for a small portion of the welcome mat population that gained eligibility under Colorado’s earlier Medicaid expansion but was not included in
the ACA expansion — namely, parents and caretakers between 60 and 68 percent of the FPL and children on continuous eligibility, meaning their Medicaid coverage remains intact even if their family experiences a change in income during the year. This analysis, however, makes the assumption that these small populations are in the standard ACA “welcome mat” group that receives its match from the state’s General Fund.
In the box titled “Medicaid: A Primer,” CHI cites eligibility guidelines before and after expansion. These eligibility levels reflect new methods of calculating income under the ACA and modified adjusted gross income (MAGI).
Because this change did not go into effect until January 1, 2014, pre-expansion regulations display the criteria for children as at or below 133 percent of the FPL and the criteria for pregnant women as at or below 185 percent of the FPL. This differs from post-expansion regulations, which display the criteria for children as at or below 147 percent of the FPL and the criteria for pregnant women as at or below 200 percent of the FPL. Despite this, the eligibility for these two groups did not actually change under the ACA expansion.
Finally, certain populations defined as welcome mat or newly eligible by HCPF were defined differently by CHI.
While foster care and non-citizen emergency populations technically include some people who are newly eligible, they are all matched at the standard 50 percent federal rate for welcome mat enrollees. For simplicity, the CHI analysis considers all spending for these populations, which is about $20 million per year, to be welcome mat spending.
Additionally, a portion of what HCPF calls the welcome mat population — parents and caretakers with annual incomes between 69 and 100 percent of the FPL — receive a 100 percent federal matching rate. CHI, for purposes of this analysis, considered these populations to be newly eligible.