Methodology Matters: Expansion Edition

by Adrianna McIntyre

Spoiler alert: it “matters” on the order of hundreds of millions of dollars.

Kaiser Family Foundation and the Urban Institute published a report last November on cost and coverage implications of the Medicaid expansion, considering both the national perspective and state-by-state analysis. If you get all the way to the bottom of the executive summary—on page one—you see the following (emphasis added):

Accounting for factors that reduce costs, states as a whole are likely to see net savings from the Medicaid expansion. Combining Medicaid costs with a conservative estimate of $18 billion in state and local nonMedicaid savings on uncompensated care, the Medicaid expansion would save states a total of $10 billion over 2013-2022, compared to the ACA without the expansion […] Net state savings are likely to be even greater because of other state fiscal gains that we could not estimate based on 50-state data.

They repeat that caveat—that there are state-level budget factors they haven’t been able to account for—at least twice more in the report’s discussion, stressing that savings realized by the states are likely to exceed their estimates. State-specific savings that KFF/the Urban Institute couldn’t account for, broken down in sections 4-6 here, can be broadly categorized as:

  • Savings from transitioning current Medicaid populations to newly eligible group, who will be subsidized by the federal government at the new, higher match rate
  • Savings from reductions in state-funded programs for the uninsured—these populations will transition into Medicaid or the exchanges, which are federally subsidized
  • Other revenue gains and savings from payers and providers

I’m not irritated by the reality that researchers often face such limitations—as long as they own and acknowledge them, like KFF and the Urban Institute did. But their report wasn’t the first time I saw this data in action—I dug it up after reading something else.

Screen Shot 2013-03-17 at 8.33.54 PMThe Heritage Foundation “disaggregated” the information from the report to provide state-level projections of costs and savings. Rather than disclosing the Urban Institute/KFF’s repeated hedges about projections missing some sources of savings, Heritage suggested the speculative data might actually overstate savings.

When I looked at the results for my home state, Michigan, I was perplexed. See, I wrote a paper last term on why Michigan should expand Medicaid, including a very compelling economic argument that exists. This confusion is what drove me to parse the original KFF/Urban Institute report—and then go back to my sources for that policy memo.

The Center for Healthcare Research & Transformation (CHRT), a non-profit partnership between the University of Michigan and Blue Cross Blue Shield of Michigan, conducted a really excellent analysis of the Medicaid expansion’s economic implications for the state. CHRT was able to account for those Michigan-specific budget offsets that KFF and the Urban Institute did not. So how much does that matter? It turns out, quite a lot. “Hundreds of millions of dollars” a lot.

I pulled cost and budget offset (savings) numbers from page five of CHRT’s issue brief. In the graph on the left, you’ll notice that my red line—gross costs—closely mimics the Heritage graph. But our “savings” lines tell vastly different stories. The Heritage graph is missing over $250 million dollars in Michigan budget offsets each year. I also wanted to show the net effect of subtracting costs from savings; net savings are visualized on the graph with the blue line. It’s true that annual costs will eventually outpace annual savings (it’s labeled and everything). But the cumulative effect over the 2014-2023 window, when you account for state-specific factors, is $983 million in savings—that’s a pretty far cry from the $1.3 billion in costs that Heritage estimated. 

Screen Shot 2013-03-16 at 9.27.03 PM Screen Shot 2013-03-18 at 9.11.38 AM

I don’t have the time to undertake this sort of analysis for every state projected to see increased costs based on the limited data used by Heritage and the Urban Institute/KFF (but if you do, check out this repository of state-level fiscal analyses). It would be irresponsible for me to blindly suggest that those other states would see savings reflective of Michigan—but wouldn’t it be similarly irresponsible if I failed to demonstrate that due diligence of this sort matters?

Adrianna is a graduate student in public policy and public health at the University of Michigan.
Follow her on Twitter @onceuponA or subscribe to the blog.


One thought on “Methodology Matters: Expansion Edition

  1. Eileen Ellis says:

    Excellent work. Michigan’s State Budget Office estimates that savings in the early years would be sufficient to cover net costs through 2035. The Senate Fiscal Agency doesn’t see savings quite that far our, but still for many years. You’re right not to extrapolate. In particular, many states don’t spend as much on non-Medicaid mental health services as Michigan.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: