In the blowback to Francis Collins’ comments about budget cuts delaying an Ebola vaccine, there is a lot of confusion going around about just how much the NIH budget declined.
The worst offender is the usually very good Sarah Kliff at Vox.com, who writes:
The NIH’s budget rose rapidly during the early 2000s, growing from $17 billion in 2000 to a peak of $31 billion in 2010. This meant more money for everything…
Funding then began to decline in 2010 and has continued to fall slightly over the past four years (this was during a period when Obama was in the White House, Democrats controlled the Senate, and Republicans controlled the House). By 2013, funding was down to $29.3 billion. These figures do not account for inflation.
Inflation – there’s the rub. Because when you do account for inflation, you see that the NIH budget was in decline long before 2010 – in fact things started to go south after 2004, as the AAAS budget analysis shows:
And depending on how you make the inflation adjustment, things can look even worse – you hear claims of a 20% decline tossed around. To understand how this works, lets look at the numbers themselves:
First, lets go to the NIH Office of Budget and look at the actual, non-adjusted NIH budget numbers.
In 2004 it was $28,036,627.
In 2013, it was $29,315,822. (If I’m reading the footnote in the table correctly (PDF), this includes the effects of sequestration.)
So in nominal terms, the 2013 budget is almost 5% more than in 2004.
Now let’s account for inflation. The NIH gives two price indices for adjusting the budget:
The standard one:
Annual budget formulation guidance from the Office of Management and Budget (OMB) typically makes reference to the Price Index for the Gross Domestic Product (GDP). This measures the annual percentage price increase for all goods and services produced by labor and property located in the United States.
And there’s a research-specific one, which is a bit controversial, based on the cost of goods and services research:
The annual change in the Biomedical Research and Development Price Index (BRDPI) indicates how much the NIH budget must change to maintain purchasing power. The BRDPI was developed and is updated annually by the Bureau of Economic Analysis (BEA), Department of Commerce under an interagency agreement with the NIH.
There are pros and cons for each one, but that doesn’t matter – let’s just look at the effects of both. We’ll convert the 2004 number to 2013 dollars. The index number is (in the 2004 column, pick the 2013 row) 120.0 for the GDP Price Index, and 132.9 for the BRDPI. So in 2013 dollars the 2004 budget was equivalent to:
$33.6 billion if you use the GDP Price Index
$37.2 billion with the BRDPI
Therefore, in terms of real dollars, the 2013 NIH budget is either 13% or 21% below the 2004 budget. Neither number is a “slight” decline. The 2010 budget was $32.9 or $33.2 billion, either 2% or 11% below 2004.
The Finch and Pea 
Great post! Thanks for crunching the numbers and doing the analysis!
It kills me when people downplay the scale of the NIH budget crunch. So maybe we wouldn’t have an Ebola vaccine anyway, but the budget really is in bad shape.
Similar trends in Canada (although decline didn’t start till 2007): https://twitter.com/jwoodgett/status/524189326502920192
The Canadian data also hide effect of flow-through/ear-marked funding for new programs such as CERC.
The hard thing is that the trends don’t capture the whole story. Science research and training happen on long time-scales. When you ramp up spending quickly and then let inflation erode the budget away, you’re taking a wrecking ball to the built up research capacity – labor market/research institutions don’t respond (and are unable to respond) very quickly.
Hindsight is 20:20 but I seem to remember that NIH was wary of the ramp up/stimulus, knowing it could result in a hard landing. Moreover, universities quickly absorbed the new funding by expanding at every opportunity (recruiting and building). The inertia led to inevitable overruns and additional mouths to feed. As I understand it, the problem was exacerbated by funding rules (NIH had few tools to mitigate the growth) as well as unrealistic assumptions that growth would continue. It is also a result of politics where promises such as “doubling” make headlines while “sustaining” suit no one.
You’re right that there were people prescient enough to be wary of such a rapid budget increase. I have the references around somewhere, but can’t find them right now.
In terms of what we should have expected after the doubling was complete, I don’t think it was unrealistic to expect modest growth – otherwise why even bother doubling? I’m not sure anyone was expecting that the budget would fall relative to inflation, and so quickly – years before the Great Recession. To use a metaphor I’ve used before, that’s like making an expensive renovation to your house, and then letting your investment go to seed.
I would argue that much of the increase had been wasted or spent in bad science. I have seen the same thing happen in the past in physics (in the field of high-Tc superconductivity), where budget increase led to too much activity, too many bad papers, but no real science. Once the wasteful scientists move out, science returns to its root and that is very healthy for the long term.
So, I consider the budget cut a blessing in disguise. In fact, it appears that there is no real cut yet, and I expect plenty to come. NIH budget is going back to its 1995 level (in inflation-unadjusted number) along with US economy.
I disagree that much has been wasted… well, leaving aside certain types of consortium science… Paylines are so tight at the R01 level, to the point that funding good grants has become a lottery – reviewers are being asked to make distinctions that are too fine, and people are wasting their time tweaking and resubmitting perfectly good proposals.
The problem is not merely the erosion of the budget; it’s that it happened after a rapid doubling. The build up from that doubling is now being wasted.
The doubling was a mistake. A commitment to slower, but consistent growth is much better than big swings.
“A commitment to slower, but consistent growth is much better than big swings.”
No. A commitment to never getting Federal government involved again is the only solution.
Scientists survived and did phenomenal work long before Fed gov started its big ramp under the excuse of fighting war against cancer. .