[NOTE ADDED: Unfortunately, within two months of this post, Mr. Zakaria was suspended from his job for plagiarism. Such a spectacular lack of integrity calls into question everything he has ever written, and so I cannot anymore recommend his article, nor will he ever be quoted on this website again.]
Today I’d like to call your attention to an article by Fareed Zakaria, entitled “How government funding of science rewards U.S. taxpayers.” (The sentiment also applies to taxpayers elsewhere, of course.) I can’t vouch for the details inside the article, but the point that Zakaria makes is one that I personally feel is very important.
When I give public talks about the fundamental research that I or my colleagues are doing, I am often asked, “what are its benefits to society?” It’s a completely fair question, but with fundamental research it is typically far too early to know the answer; it can be many decades before the benefits, if any, become evident. I think the best answer requires a long view — the kind of view Zakaria lays out in the article. I often reply this way: that you should think about government investment in fundamental scientific research as similar to venture capital investments in many small startup companies; most of these efforts will fail, or will succeed with a small payout, but one or two will pay off in spectacular fashion and change the world.
And you surely want that payout to happen in a friendly country. Zakaria points out the worrying slope that the United States is on; though scientific breakthroughs have a big impact on the economy over the long term, funding for science is on a long-term decline (as a fraction of GDP) in the United States, while it is sharply increasing in a list of countries that include some that are not friendly to the United States.
Zakaria focuses on what is happening today in biotechnology, genetics, genomics, etc. He also mentions the historical case of the transistor, the device that lies at the heart of our computer-based society. This last is an even nicer example if you expand your view. The research that was done in the late years of the 19th century on the emission of light by atoms and on the electron led eventually to the equations of quantum mechanics, which in turn were essential in the development of the transistor. No 19th century scientist could have predicted that the discovery of the electron would help put a cell phone in your pocket.
[Thanks are due to Leonid Kruglyak for bringing this article to my attention.]
18 Responses
Oops, make that [[Matthew Strassler]].
For those who are fans of govt. support of scientific research, one small thing to do might be to create a Strassler wikipedia entry. Many physics bloggers, e.g. Lubos Motl, Sean Carroll, Clifford Johnson, Peter Woit, Jacques Distler, are already in wikipedia. Go to
http://en.wikipedia.org/wiki/Seiberg_duality and create a Strassler entry using [[Matthew Srassler]]. Merely log in and imitate other physics stub entries.
Can I only make a new entry or can I remove one from the list, that should not be there, too … 😛 ?
Prof. Strassler of course belongs to the list on nice bloggers, good idea 🙂
…similar to venture capital investments in many small startup companies; most of these efforts will fail…
I think it’s hard to make the analogy that the LHC is a “small startup company”. If the idea is to fund lots of small and varied projects, wouldn’t it make better sense to spread the money around to many different labs, each working on very different research? In physics terms, your could fund hundreds — if not thousands — of small independent teams of condensed-matter, chemical, biological, optical etc physicists all for the price of the LHC.
I come from a particular point of view in saying that science is too expensive for the public to fund without clear justification, which is difficult when we need funding to test hypotheses that by definition carry risk of failure. It all becomes peicemeal unless there is allocation across a spectrum according to a complete overview of science. However, there is no overview because science is not yet unified in theory, for allocation, and we are left with competing claimants. Allocation becomes a scientific process in itself.
Until science has greater unity and therefore capacity to value and prioritize between alternatives, I would fund social services first and science a distant second. Thankfully, I have recently unified science in my book at http://www.thehumandesign.net (its a non spiritual geometrical Design to unify the four forces). Sadly, I am not a scientist, so even the unification is not really yours, it belongs to a lay person. That speaks to the problems of current science.
Stop the pipe dream nonsense, human spaceflight program, and you will have plenty of research money to cover all sciences across the board.
The problem started with Congress by including the word “manned” in one little paragraph in the NASA Act of 1958 and the Texans ran with it. President Kennedy didn’t help either.
Matt, I’m sympathetic to what you write here, but how do we translate it into a workable principal – in other words, should federal funding for basic science be 10x as large? 100x as large? Exactly as is? Half as large? It may well be that the average dollar spent on science has an enormous return but to answer these questions we need to say something about the return to the marginal dollar.
Your answer seems to be, “Well, we can never know what that is, because when we calculate it we always miss the small probabilities of an unforeseen enormous return in the distant future.” Fair enough, but the distant future is also typically discounted heavily relative to the present for several reasons in cost-benefit calculations (among others, the fact that growth and technological progress means that in expectation people are poorer and needier now than they will be in 100 years). So while I accept your judgment that investment in basic science can have enormous very long-term benefits, I don’t know how to go from there to saying, “We should invest more in basic science.”
It’s a very fair point. Evaluating how much is the right amount to invest is just as difficult for the venture capitalist. But in making the evaluation one must understand that the benefits are not something that you evaluate by asking “what good is it this knowledge for current technology?” You really have to ask “do I have a diverse portfolio that gives me a reasonable chance of making a killing?”
Just to play devil’s advocate, isn’t it safe to say that whatever is discovered at the LHC and beyond will not have any practical effect until our everyday technology involves TeV scale energies? Anything involving lower energies should be described just fine by the Standard Model. In other words, doesn’t the paradigm of effective field theory actually make high energy physics irrelevant to low energy everyday life?
I don’t know if it is safe or not. Various physicists have embarrassed themselves in the past by stating that a certain new discovery would never be useful for anything.
I am not sure what kind of practical technological advances can come out of studying the Higgs physics and the like, in the same way that the transistor came out of the discovery of the electron and quantum mechanics. The only reason I can see of for doing high-energy physics research is for the sake of plain curiosity and the fact that the state of understanding and technology allows for it.
Various physicists have embarrassed themselves in the past by stating that a certain new discovery would never be useful for anything.
I completely understand this objection and I too suspect that greater knowledge of the Higgs will likely never bring us great technological advance (or at least, not in the near future). But I also suspect this is the wrong question to ask – that the benefits of pure science will in the future descend less directly from the science itself, but will still be inextricably linked to the research which bore it. Just as CERN brought us the web, and the space programme brought numerous technological spin-offs, I’m sure pure inspirational science of the future will reward us in ways we cannot now predict, but which stem directly from its pursuit (indeed, just as times of war always bring us medical advances). Apart from anything else, you can’t bracket off different areas of physics, or even science as a whole, and treat them as isolated avenues of study – the different disciplines will always talk to each other and augment each other’s development. Matt is right to note that this is a horse race in which you can’t pick the winner before it starts, but the truth is bolder than that – whatever winners there are will emerge stronger from the participation of all the others.
It is a sad thing to see how the US have carelessly thrown away the “fundamental physics ball”. But as long as there are other powerful “players” around willing to pick it up, this is not that bad for “global” science as a whole. I think in the end it is not that important which countries keep fundamental physics going but that somebody does it …
I think the transistor is a great example in this context. Everyone can appreciate the inability of 19th century scientists to imagine the transistor, but everyone can see the every day benefits that arose from that research. Speculate to accumulate as they say.
I agree 100% that supporting fundamental research is in the best interest of the general public. However, I disagree that the support has to come from governments.
Picking on your above example of venture capital:
Relying on government funding of new and creative industries does not have a proven track record. That is the whole reason the venture capital market exists. Governments have incentives to not be risky, play it safe and fund the current line of thinking. They have very little incentive to fund projects that are risky. If funds are “wasted” on a project that fails or doesn’t live up to expectations then the government officials who approved the project will be reprimanded or fired. Venture capital does not work that way. Startups convince investors that they are worth the risk. They don’t outline every single detail of what they will do and how they do it. If they new how to do it exactly and they just needed money, then they would go to a bank not the venture capital market.
It very hard to fund anything that is perceived as risky with government funds. Today, large projects are only given funds after years (or decades) of work go into designing a project to demonstrate the project is not risky. Those years could have been better spent actually building the project in the same way venture capital is given to startups before they make their product and the startups use the money to realize their idea. If it fails, it fails and the investors loose money. If it succeeds, everyone wins (except for the theorists who will have less time to churn out papers as their parameter space dwindles at a faster rate. Think about how many less SUSY papers would have been written if the SSC had be built)!
I think the future of high energy physics research funding must come from the private sector. This is the only way that breakthroughs will continue to happen. A 14 TeV LHC was not risky, why was it not 50 or 100 TeV (and why was it not built in the 80’s)? Don’t even get me started on what is happening to LBNE now. It was once dreamed to be a fantastically rich multipurpose experiment, now it will be (if it even gets built) barely able to compete with current generation experiments.
I know that government funding is more or less the only game in town for HEP research these days, but it will never change unless we, as a community try to change it. I also know I am not providing answers and just complaining but I just get so mad sometimes about the dismal future I see for experimental HEP. Maybe just thinking that an alternative funding source should exist is a good place to start. Or maybe I should finish my proposal for the experiment that will not be risky with today’s technology but will not be given government funds to build until 2025.
The issues you raise are indeed very serious. Governments are too conservative, right now, in the way they invest in the future; it’s a mistake. Your suggestion of private funding is fine in principle, but challenging.
Of course it would be great if philanthropic donations came from wealthy individuals (as has happened a couple of times in the past few years) or large foundations (as do often contribute to, say, telescopes) to make this type of scientific research possible. But can we really fund projects in the $100 million or $1 billion range this way, especially on a regular basis?
The alternative would be to really do something like venture capital, where the investors who put in $100s of millions can potentially make billions off the science — requiring there be patents covering anything that we happen to find (or develop along the way) that has a practical use. But then you would be getting into the very questions that the human genome project ran into (can you patent a gene that you’ve discovered? could you patent a new finding on neutrinos that made it possible for you to use them, say, in searching for oil?) If you can’t patent the discovery, then anyone who doesn’t have to invest the money up front is better off waiting for the discovery and then going off to patent an application that they develop without making that initial investment — and so no one will invest. I’m oversimplifying but the problem should be clear.
The only investors who can make money off the science without holding a patent are taxpayers themselves… who by each putting in an average of $100 into a few scientific areas may perhaps find that their children’s cost of communication consequently dropped by $30 per month, or that their children’s quality of life is improved by all sorts of job opportunities that are hard to put a $ sign on, etc., etc.
Is there a good middle ground between the venture capitalist and the taxpayer?
I’d have to disagree about risk appetite. The private sector has a very high risk premium. Meaning, they expect to be highly compensated, quickly, for any risky investments they make. If you look at PE or VC projections, they want to make 20-50% annualized returns on their investment. How much research can really claim that sort of return even in best-case scenarios? A lot of fundamental research has payoffs that might be decades in the making (and hard to monetize), and there is simply no investor who is willing to bear risk that long.
Government is the ideal source of funds for the sort of research with long-term, perhaps indirect, benefits. Governments have very large budgets and so can bear risk through diversification. They also have no need to directly monetize the research–the payout is the benefit to society, without a need to convert it back into dollars.
It may be true that not enough worthy, risky research is getting funded, but I think that’s a problem of budget and not a problem with a taxpayer-funded model. If the budget is too small, capacity for risk through diversification is reduced. Not to mention, it’s very expensive to *evaluate* high-risk projects. If the budget is strained, one of the first responses will be to be more conservative with the available funding, and perhaps that’s what’s happened.