Stimulus Spending and NSF Funding

Both the economic stimulus plan passed by the House and the plan passed by the Senate have quite a bit of funding for scientific research. As most of you know, scientific research is near and dear to my heart, and I'm generally in favor of spending lots of money trying to learn new things. For today, though, I'm going to set aside my usual criteria for evaluating scientific spending (more money = more better), and look at some of these proposals in terms of their effect as economic stimulus.

There are quite a few different sections of the stimulus that contain funding for science. For now, I'm going to just focus on the National Science Foundation's share of the stimulus. Under the version of the stimulus passed by the House (pdf), there's an additional $2.5 billion for NSF's "Research and Related Activities" budget line. $200 million of that is specifically for "research facilities modernization", and $300 million is reserved for big-ticket projects funded through the Major Research Instrumentation Program. The only remaining restriction is that peer-reviewed research projects funded with this money need to be selected within 120 days of the bill becoming law. Under the version of the bill currently being considered in the Senate, there's $1.2 billion for the "Research and Related Activities" budget line. There are no specific projects or spending limitations specified in the Senate version of the bill.

In terms of scale, this represents a 20-50% increase in funding for this budget line, depending on which version of the bill is ultimately passed. (Yes, I am assuming that the final version contains some version of this funding.)

If the majority of this money is spent in the form of research grants, it will be dolled out in relatively small amounts (the median grant is about $100,000, spent over about 2.5 years). It will be spread out to institutions all over the country. A small part of each grant will go to cover overhead costs at the receiving institution, with the rest of the money being spent on salaries and on other research expenses. (Just what the "other research expenses" are will vary wildly from proposal to proposal.)

Most of this money can be committed fairly quickly. At the moment, NSF is funding about 25% of the proposals they receive. They are not funding at that level because 75% of the proposals they get suck. They're funding what they can with the money they have. The additional funding will let them bring the funding percentage up to about 30%, at least for a little while.

That's good for science. But is all of this stimulus? As much as I'd love to be able to say, "Of course it is!" the situation isn't as clear cut as it is with some of the other proposals in the bill.

In terms of job creation, this probably isn't going to do a lot. Some soft money researchers will get to keep their jobs. A bunch more researchers will be able to tell lab techs that they've got job security for another couple of years. More grad students will be able to concentrate on their education instead of worrying about the rent. All of that is good - particularly the part that has scientists eating actual food and sleeping under actual roofs - but I'd be surprised if the total number of jobs directly saved through this funding exceeds 25,000.

In terms of jobs indirectly saved through this money, I don't know how you could even go about estimating that. This money is all going to be spent, but the money that's not spent on payroll is going to be spent on buying acetone, funding research trips to Zimbabwe, and everything in between. Lots of money is going to be spent, but its going to be spent on many different things spread across many different areas of the economy.

Then there's the indirect, long-term stimulatory effects. One of the reasons that it makes sense for the government to fund basic research is that this is not usually research that leads to the sort of breakthroughs that create major and immediate economic payoffs. That doesn't mean that it has no economic benefit, or no long term potential payoffs. Some of this research will probably be pretty unproductive. Some of it will probably provide the basis for the kind of research that does wind up leading to short term payoffs. And no, we don't really know which will be which ahead of time.

As Einstein allegedly once said, "If we knew what we were doing, it wouldn't be research."

This is money that I think is in our national interest to spend, for lots of different reasons. Ultimately, though, I think it's hard to say that this money is going to be as stimulatory as the weatherization project I talked about last week, or even as stimulatory as increasing funding for food stamps would be.

Am I wrong about this?