randombio.com | science commentary
Friday, September 16, 2016

Privatize science research. Here's how.

American science funding is unsustainable. It must be freed from its dependence on government.


W hether people really want new technologies, knowledge about the universe, and cures for diseases, or whether they just think they do, is debatable. But most would acknowledge that scientific advancement is important for our long-term economic survival. What is missing is a viable business model that can make it possible.

There's no denying it: science research is now utterly dependent on government funding. Whether they realize it or not, grant-funded researchers are de facto government employees. This is the cause of most of the problems that science is facing.

Current funding levels are unsustainable.

Unless radical changes are made, government funding will dry up over the next few decades. We must start preparing now.

Funding for scientific research is discretionary. It is caught between the rock of ever-expanding entitlements and the hard place of economic stagnation.

In 2015 the federal government spent 66.6bn on non-defense R&D. By contrast, Medicare alone is over ten times higher; non-discretionary spending, interest on the debt, and military are 89% of the budget. Who can doubt what voters will select when faced with a choice between cutting Medicare or Social Security and cutting scientific research?

Already, the East and Southeast Asia region has become the biggest driver of science funding (36.8% vs 29.4% for North America and 21.9% for Europe[2]). This year China surpassed the United States in Nature's index of high quality research publications.

Government funding compromises the integrity of science.

When government pays for research, they can (and do) dictate what results are acceptable. When a researcher writes a grant, the top consideration is not “Is this good science?” but “Is this something the government would be interested in funding?” It's true that most politicians have no clue about science, and they defer to scientists about the details of conducting research. But things like the dietary cholesterol scandal show how pre-defined conclusions imposed by funding agencies can skew research for decades.

Many of the problems in science are caused by by government giving researchers a message, sometimes subtly, about what conclusions it wants. Just to give one example: regardless of your position on global warming, all researchers know which side of the argument government is willing to subsidize and which side will be starved for funding.

Then there are the activists. Activists and special interest groups lobby furiously for research earmarked for their pet causes. The result is that funds are allocated on the basis of hysterical news reporting as much as scientific merit. Funding agencies try to maintain a balance, but there are limits to what they can do.

In the struggle for funding, scientists are tempted to do science by press release: providing grandiose claims to the news media for the sole purpose of generating excitement during the next round of grant review. As a result, the public grows suspicious of all science: if there are so many breakthroughs, why are there never any cures?

To preserve the integrity and economic viability of science, it must be funded from an independent, unbiased source.

Science employment statistics are bad and getting worse.

An even bigger scandal is in the horrible employment numbers for postdocs. Their dire situation is laid out in a report by the National Academies of Science, Engineering, and Medicine. B.L. Bernderly at Science Magazine writes:

For all but a small percentage of aspiring researchers, doing a postdoc at a university is a lousy idea because it will neither result in an academic job nor otherwise advance one's career. If graduate students had accurate information about what lay ahead, many would—and should—choose not to become postdocs.

When students discover they're being tricked, they'll avoid careers in science. The H-1B program, which has devastated science employment, would expand to fill the gap, perpetuating a vicious cycle of dependence and decline.

Billions are wasted on pet projects by well connected senators who try to funnel tax money into their states. More billions are wasted on projects that, while potentially important, have little prospect for scientific advancement. Who pays the piper calls the tune, and right now we're all humming a catchy little ditty that sounds a lot like the national anthem.

History shows that science advances fastest when scientists are free to study scientific puzzles. When they are, we get penicillin and the theory of relativity. When they're not, we get the H-bomb. All these things are useful, but one path leads to more scientific advancement and the other does not.

Funding stinks.

Only tenured professors at universities get the beautiful women, champagne and free Maseratis. Their job is to teach, attend faculty meetings, and write grants. Their students and postdocs do most of the actual work.

The rest are research faculty who are on soft money, which means the institution gives them a desk, an empty lab, and some electric outlets, but doesn't pay for them. Unlike employees with regular jobs, scientists are expected to find a way to get the equipment they need and to find somebody (i.e., the government) to pay their salaries.

Scientists actually pay their institution for the privilege of doing research by including ‘overhead’ aka ‘indirect costs’—which can be as much as 90%—tacked on to the grant's budget. When a grant comes in much of that goes straight into salaries of the bureaucrats and administrators who now dominate the employment rolls at the universities.

Research scientists also have little job security; if their grant runs out they are basically out of a job. NIH funding levels are hovering around 11% for first submissions and 18% overall. So, like their tenured comrades, they spend far more time writing grants than doing actual research.

This may explain why many scientists, despite being libertarians at heart, have transmogrified into advocates for bigger government. I've seen sober-minded Ph.Ds get thrown into a panic at just the suggestion that some politician is threatening to cut funding for science.

It also explains why we write 2.2 million new science papers per year[1], yet we still don't understand what causes heart disease and we still haven't found cures for cancer or neurodegenerative diseases. We are feeding off the taxpayer and it has corrupted us.

Here is my proposal.

When I worked at a nonprofit research institute, I recommended to the boss that they should not spend the startup money they received but invest it and run the institute from the returns.

Yeah, well, nobody ever listens to me. They spent like there was no tomorrow. They hired people they couldn't pay for. They bought equipment they didn't need. They spent millions on things like patent lawyers and new ceramic floor tile. And they quickly went bankrupt.

This happens not because lab directors are greedy but because they know the money could disappear and reappear at the whim of some politician. This is crazy.

We've all seen those sci-fi movies where somebody leaves 27 cents in their bank account and wakes up a million years later to discover that they now own 7/8 of the world's economy. It sounds like a joke, but the principle is valid. The reason government isn't funded that way is not that it's not practical. It's because taxing and spending gives you control, and control, not economic sense, is what government is all about.

Big institutions like Harvard Medical School work this way already. Much of their spending comes from their endowment, which liberates them from political influence. It is one reason they are so successful.

Here is the plan:

  1. Create a research fund.

    Ten percent of all government grants should go into a general fund, sort of like a giant 401k. The principal should remain untouched and half of the returns should be available to researchers, using the same scientific peer review system that the National Institutes of Health uses now.
  2. Eliminate paid journals.

    Researchers never get a cent for writing or peer-reviewing articles. They'd be reading the ones they review anyway; the only difference is that when reviewing them they write their critique down. But one of the biggest impediments for a freelance research institute is the enormous cost of journals.

    Journals aren't like magazines. Libraries are charged thousands per year for each journal subscription, and there are thousands of journals. Fee-based services exist, but even these would quickly bankrupt a small organization. To add insult to injury, journals now demand payment from authors of hundreds of dollars for the privilege of publishing their research.

    All new peer-reviewed journal articles should be posted on a free website, like arxiv or biorxiv.
  3. Get scientists out of the universities.

    There are some advantages to being at a university—cheap graduate students being the main one—but those benefits accrue mainly to tenured faculty. There are many advantages to being far away from a university, such as lower real estate costs and fewer dumb rules.

    At universities we get endless lectures about how to sexually harass each other[3] and which pronouns we're allowed to use. Just last week instead of working I had to listen to a two-hour lecture on how to use compressed gas cylinders without getting killed, something I had somehow already managed to do for over thirty years. Then there was a mandatory four-hour radiation safety refresher course and the mandatory animal safety course and the mandatory chemical hazard safety course. Before that we were told we couldn't have any chemicals in our chemistry lab because they were dangerous. And we weren't allowed to have glass beakers on the shelves because they might fall off and hurt somebody.

    Research institutions should be located far enough away from universities to avoid this idiocy but not so far that equipment repair costs are dominated by travel time.

The transition would be gradual and painless. NIH's extramural budget is about 26bn per year. If 10% of this went into the research 401k fund, at a 10% return over inflation within 24 years interest from the fund could completely replace funding from NIH. It would double thereafter every 7.28 years, but with continually decreasing cost to the taxpayer.

The tenure system would remain intact. But it would force the universities to compete for good researchers. The current system churns out people regardless of whether there's a job market for them. This is worse than wasteful; it is cruel.

During the transition the skilled and dedicated people at the Center for Scientific Review—one of the few government organizations that is not yet hopelessly corrupt—could use the same peer review system, which has been fine-tuned for fairness over the years. Of course, if the government wanted to earmark special funds for their pet projects they still could.

But those of us who just want to advance the nation's science and health would be able to do so. Science would be liberated from the government-industrial complex. Instead of being parasitically dependent on the whims of politicians and political activists, we would be free to do our job of solving important problems.

Maybe more importantly, we would regain our scientific integrity.


1. National Science Foundation Science and Engineering Indicators 2016 nsb20161.pdf Chapter 5, page 7

2. National Science Foundation Science and Engineering Indicators 2016 nsb20161.pdf page O-15.

3. Or maybe it was about how not to harass each other. I don't remember. I was sleeping through most of it.


Last edited Sep 17, 2016 4:29 am

Related Articles

Privatizing science research, part 2
American science funding is unsustain­able. It must be freed from its dependence on govern­ment.


On the Internet, no one can tell whether you're a dolphin or a porpoise

back
book reviews
home