In the New York Times today William J. Broad reports (registration required) on Kei Koizumi’s presentation at AAAS on R&D budgets that we referred to yesterday. The Times summarizes the implications of Koizumi’s analysis as follows,

“Federal support for research and development stands at $126.5 billion this year, and the administration has proposed increasing it over five years to $141.6 billion. But Mr. Koizumi found that large projected increases for research at the Department of Homeland Security and the National Aeronautics and Space Administration masked steep declines at all other nondefense agencies.

For instance, he said, federal budgets would decline 15.9 percent for earth science over the next five years, 16.2 percent for aeronautics, 11.8 percent for biological and physics research, 21 percent for energy-supply research, and 11.3 percent for agriculture research. Research budgets would drop 15 percent at the Environmental Projection Agency, 10.5 percent at the National Oceanic and Atmospheric Administration and 4.7 percent at the National Science Foundation.”

Koizumi’s analysis can be found here.

Budgets are slippery things to get a grasp on, so some perspective is worth providing about the AAAS analysis and Times report.

For the sake of discussion, let’s assume that Koizumi’s projections turn out to be on target (about which I should note Koizumi observes, “It’s only one idea of the future. But I show these because it’s an important consequence of the deficit. The president’s budget proposed tough choices.”)

It turns out that because 2005 happens to the all-time high water mark for overall federal funding for R&D, as well as for most agencies, any future reduction in budgets will look large as a percentage. But an accurate understanding of budget reductions requires placing them into the context of projections in the overall federal budget. Current projections have non-defense discretionary budgets returning to 2002 levels in 2009 (in constant dollars). Under the AAAS analysis in 2009 the budgets for the major science agencies would return to the following historical levels (again using constant dollars):

DOD all time high
DHS all time high
NASA highest since Apollo
NIH return to 2005 levels
NSF all time high
USDA return to 2004 levels
Interior return to 2002 levels
USDA return to 2002 levels
DOT return to 2001 levels
EPA return to 2003 levels
DOC return to 2001 levels

The bottom line: If overall non-defense discretionary funding is reduced in real terms from 2005-2009 to about the equivalent of 2002 levels, while the falling tide would lower all boats, it appears that with the exceptions of DOT and DOC, federal research and development agencies do no worse than the average decline, and in some case do significantly better.

Perhaps this is one factor underlying an observation made by Senator Tom Daschle’s (D-SD) in his AAAS speech yesterday, “But we should be honest with ourselves. Outside the scientific community, there is no hue and cry for more government funding of R&D.”

More generally, if the scientific community really wants to justify that it should receive a disproportionate share of federal funding as compared with the multitude other demands for public resources, shouldn’t the discussion about science and technology expand beyond discussing only the size of budgets? Again Senator Daschle, “We have not done enough to show the American people the connection between the work underway in your laboratories and the problems that affect their lives… The challenge to the American scientific community is to rebuild the link not only between science and government, but between science and society.”