Higher Education Strategy Associates

Tag Archives: Basic Research

October 12

The Right Way to Argue for Basic Research

The week before last, you may recall, I took issue with the way the country’s illustrious top university presidents (Gerforno, for short) were trying to sell higher education.  Effectively, what they were doing was selling higher education’s research mission by claiming “look, basic research creates jobs” on the basis of a few anecdotes.

The feedback I got was mostly “we really like the portmanteau Gerforno but are not necessarily convinced that there’s any other way to argue for basic research – if all the government wants to hear about is jobs, growth and middle-class families, isn’t that they way Presidents have to sell it?”  And there’s something to this view.  I used to work at Universities Canada; I get their lobbying strategy, and yes, this is exactly what the Presidents are thinking.

My objection is effectively two-fold.  Part of it is that there is too much intellectual dishonesty involved.  While there is a lot of research that ties a country’s overall research spend to long-term GDP growth rates or productivity rates, there is very little to substantiate a link between publicly-funded research or basic research with same.  That’s not to say the link doesn’t exist, just that the research doesn’t exist to prove it.  And second of all, there’s frankly a risk that coming to government year after year and saying “research means higher productivity means more jobs wears thin after awhile.  I mean, it’s been 20 years since the feds created the Canada Foundation for Innovation and really started spending money on research and as far as I know no one’s suggesting that all that money has in fact moved the needle on productivity very much.  Now, if we re-ran the last twenty years to answer the counter-factual “and what would productivity and growth look like if we hadn’t spent all that money” we might find it has in fact been worth it, but it’s easy to understand why politicians and civil servants might be skeptical about this claim.

So now many of you are now thinking “ok smart guy, how would you argue for it”?  Fair challenge.  Here’s my take:

I agree that there needs to be a way to link research to the idea of prosperity, but I think it has to be done in a way which is at once both much more specific and much more general than the way we are doing it now.  More specific in the sense that we have to stop arguing that research on its own is going to deliver growth because that’s a completely nonsensical proposition.  Investment in research is a necessary but insufficient condition.  To credibly link research to the economic engine, the research agenda needs to be tied to a whole bunch of other agendas: a competition agenda, a tax agenda, a regulation agenda, etc.  And the higher education agenda needs to make allies among people in other sectors who can help broaden this agenda.

At the same time, research advocates need to be less specific in the sense that we need to stop claiming that investment in this or that particular new shiny thing is going to lead to breakthroughs/growth/prosperity.  Every time higher ed leaders try to make the research-growth link, they reach for a small handful of specific examples in the life sciences or ICT.  Now maybe those leaders do actually know better and are just cynically pandering to politicians who have a narrow idea of what productivity growth actually is (“It’s new! It’s shiny!  Therefore it must be making Canada more productive!”).  The problem is that even though higher ed leaders are making these arguments in favour of “research” broadly defined, what politicians and policymakers tend to hear is “hey, we should double down on life sciences and ICT”.  Which is pretty much the opposite of what most of the research community wants.

So here’s my pitch: what we want and need is a Smart Canada.  We have no idea what the next big thing is going to be, nor does anyone else.  But our best bet to get ahead of the game is to be at or near the technological frontier in as many fields as possible.  And that means two things: first, it means adopting a very forward-looking posture on digital infrastructure and policy.  I don’t want to bore you with details, but we should be looking to imitate places like Estonia, South Korea and Singapore, and remembering that investing in digital infrastructure & digital public services > investing in digital companies.

Second, it means investing more in supporting and attracting talent.  And by talent I mean primarily people who can expand our country’s capacity in both research and development.  On the development side – that is, mostly in the private sector – that means a whole raft of changes to immigration (Dominic Barton made a number of helpful suggestions in this regard last year).  On the research side – mostly in academia – it means ensuring that we are creating an eco-system that can sustain basic research.  By and large, this means a greater emphasis on i) funding ideas and researchers rather than building infrastructure and ii) spreading the money around more widely than is currently the fashion.

(and yeah, there’s still an issue about ensuring ideas from basic research actually  do find their way into the economy eventually – that problem doesn’t go away.  But it’s kind of a last-mile problem, one you fix after the other pieces are in place).

Basically: support digital, support talent, support researchers and avoid confusing innovation policy with regional development or industrial strategies.  We can’t know for sure how to profit from tomorrow’s technologies, but this is a sure way for us to be as close as possible to the front of the line on new technology.  Think of it as an insurance policy, or if you’ll forgive a particularly ugly phrase, a way of “future-proofing” the country.

I admit it’s not quite as simple an equation as the Shiny Things = Growth algorithm which seems to have entranced the federal innovation ministry and in the short run it may be a tough sell.  But it’s a more durable and ultimately inclusive formula for linking growth and research.  We should give something like it a try.

September 26

Arguing for Science in All the Wrong Ways

You can tell it’s pre-budget consultation time in Ottawa because university Presidents are writing op-eds about the importance of research and backing the Naylor Report.  But man, are they ever unconvincing.

Let’s start with University of Toronto President Meric Gertler’s September 12th Toronto Star op-ed entitled “Don’t Let the World Pass Us By on Science”.  The sentiment is fine, I suppose, but the specific evidence Gertler uses to back up his claim is – to put it politely – weak.  It says that we are falling behind countries like “China, Switzerland and Singapore”; the evidence for this is that their universities are bouncing up various international rankings tables while Canadian universities are just staying stable.

Three points here.  First, the idea that Canada needs to spend billions on research so that U of T and a handful of other institutions can be rankings big-shots is perhaps the worst possible argument for supporting the Naylor report.  Second: “China, Switzerland and Singapore”?  What kind of trio is that?  If those three – which between them don’t have more than a dozen genuinely world-class universities – are the only ones we are worrying about, doesn’t that imply that in fact we are doing reasonably well compared to the traditional powerhouses like the US, UK and Germany?  Third, Gertler’s assertion that these three countries are “making aggressive investments in scientific infrastructure and researchers” just isn’t borne out by the data, at least over the last four or five years: in none of those countries are university budgets keeping pace with inflation and growth in student numbers (I’ll be showing this in more detail in a series of blogs over the next few weeks – stay tuned).

The second article of note saw Gertler team up with McGill’s Suzanne Fortier and UBC’s Santa Ono (let’s portmanteau them collectively as “Gerforno” to keep things simple) to pen an op-ed for the Globe and Mail called Ottawa must improve research funding – or risk losing the innovation race.  This one was a bit better as a pitch: now the rationale for investment is Canadian prosperity rather than U-15 bragging rights.  But the logic underpinning it is deeply problematic and points to the continuing poverty of the Canadian innovation debate.

The Gerforno argument runs like this: “hey, look at the breakthroughs Professors at our three fantastic universities have made: Geoffrey Hinton (artificial intelligence, U of T), Bernard Belleau (molecular pharmacology, McGill) and Michael Smith (biotech, UBC).  Their brilliant work in basic science led to huge advances that turned into big companies and lots of jobs, etc.  See?  Basic researcher equals innovation equals Canadian prosperity!”

Now, it’s definitely gutsy of Gerforno to make this argument on national economic grounds when both Hinton and Belleau sold their companies to foreigners (Google and Shire Pharmaceuticals) and Smith chose to found his company in Seattle.  But leaving that aside, as an argument this is still just arguing from anecdote. It doesn’t even attempt to draw conclusive links between basic science and economic growth.

Canada is a tiny country.  In terms of population, 0.5% of the world, in terms of GDP maybe 2%, in terms of science maybe 4%.  The firms that employ Canadians import – minimum – 95% of the technology with which our firms work.  Adding a billion or two to university research does not change that.  We will always be takers of technology, and – as importantly – most of the economic impact of ideas generated by Canadian scientists are going to occur outside our borders.  That’s how open science works in an open economy.

The effect of that extra billion or so on national levels of productivity is going to be tiny.  Sure, we might be able to generate a few more hi-tech or life-sciences companies, but that’s if and only if the rest of the innovation infrastructure – venture capital, skilled managers able to take companies from start-up to production to IPO, intellectual property regimes, tax structures, legislation limiting non-competes, etc – is all working as well.  And I think there’s a fair bit of evidence to suggest that all of this isn’t in working order.

At the national level, research capacity is a necessary but insufficient condition for innovation.  Universities, though, have taken to pretending that this is not true, that is sufficient and that a pot of gold lies at the end of the rainbow, if only we can get CIHR and NSERC budgets up another few percent.   They therefore have every reason to play down the possibility that more money on scientific research may be – from the point of view of innovation and economic growth – wasted.  Or that in the absence of the rest of the eco-system being in working order, more research money may be like pushing on a string.

Now, that’s not a reason on its own to dismiss the Naylor Report or the idea of increasing research budgets.  Maybe we ought to do it because science is good in and of itself and a more science-oriented society is a better society.  But if we’re going to justify science in the name of economic growth, we have to do better than arguing by anecdote and pretending that there’s some kind of straight unmediated line between research and growth.

Gerforno can do better than this.  Heck we can do better than this.  We have to.

April 12

Going Overboard on Basic Research?

I’m getting some worrying vibes from the new federal government.  It’s nothing I can directly put my finger on (other than some annoying Ministerial tweets last week which seemed to claim that any money put into PSE infrastructure is ipso facto about “innovation”) but I get the sense that the new government is in danger of making some real mistakes with respect to innovation policy.  Specifically, I’m worried that in the rush to repudiate the Harper legacy in all things science, they may end up with an innovation policy that takes us back to the naïve 1990s.

What do I mean by this?  Well, in the late 1990s, when the Chretien government began seriously investing in research (after having initially slashed the bejesus out of it in the 1995 Budget), their rationale went something like this: growth requires innovation, research is the wellspring of innovation, therefore:

      $ to universities for research → a miracle occurs → productive high-tech economic future

And on that not very sophisticated basis, billions were spent.

Now, without denying some good came from this, I think it’s fair to say that this is a pretty limited view of how innovation works.  For one thing, there’s an implicit suggestion that innovation is about “new discoveries” being turned into “new products”.  And while that is one type of innovation, it is far from the only one.  What about process innovations or business model innovations, to name but two?  Why focus on the “big breakthroughs” when so many incremental innovations are possible?  Why focus on only one part of the value chain (and possibly not a part Canada is particularly good at) when there is value in so many others?

To put this more bluntly, to assume that basic research is the only type of research an innovation policy should fund is crazy.  Serious countries understand this.  It’s the reason, for instance, that Germany, besides funding its universities and the Max Planck institute, also funds the Frauenhofer Institutes, which is one of the world’s greatest performers of applied research.

Over the course of the last few years there have been many complaints that the Harper government focused too much on applied research.  True, all granting councils (but especially CIHR) were pulled in the direction of having grantees justify their funding in terms of “immediate benefit” and finding commercial co-partners, etc, and for the most part this idea of injecting some “appliedness” into basic research funding was bad policy.  But the fact is that the actual amount we spend on research which is exclusively applied in nature – that is, Frauenhofer type-stuff, or programs like the Industrial Research Chairs – is actually pretty small.  The revamp of the National Research Council was a stab in a Frauenhofer direction – albeit a somewhat clumsy stab, with over-inflated expectations of quick success.  But now even that’s been thrown into question, the revamp now “suspended” pending the outcome of a review of the government’s review of its basic science policies. 

To be clear, it’s not that the government has yet made any definitive false steps.  But rhetorically it seems to be backing itself into a corner in terms of thinking of innovation exclusively in terms of basic research plus maybe funding some exciting business/university co-location spaces (an idea which I think we could also describe as being less-than-fully-baked, as I explained back here.  That would be a bad mistake.  What Canada needs is a full-spectrum innovation policy, one which doesn’t put all its eggs in the new discoveries/new products basket. 

Or, to put it another way: yes to basic research, but stools need more than one leg.

October 23

Where the Questions Are

I had planned to continue on today with my series about operating budgets by taking a look at some scenarios for Central Canada, but I’ve been on the east coast for work the past couple days, and so that post will have to wait.  We’ll get back to it shortly, I promise.  But for now, let me turn to something I’ve been thinking about lately.

One of the maddening things about many discussions that concern higher education and business is the crudeness of many popular views on their relationship.  Mostly, we hear about how business’ role is to “contribute” to higher education, either via taxes, or philanthropy, or both (depending on where you are on the political spectrum).  Often times, the role of business is to hire “our” graduates (and if that’s not happening then let the agonized introspection begin).

And while those things are all true, what these analyses actually miss is the true role of business, particularly with respect to science: it’s a huge, incomparable reservoir of questions to be answered, and problems to be solved.  Of course, people get this at the level of applied research – by definition, when companies engage with higher education on applied research, it’s to solve specific problems – but they have trouble understanding when it comes to “pure” research.  Partly, that’s due to rhetorical confusion – the wording of “pure” research (a rhetorical device of Vannevar Bush designed to keep money flowing to universities after World War II) implies that interaction between scientists and pretty much anyone else will “contaminate” research.

But a quick history of 20th century science will show you that this is nonsense.  Much of Einstein’s early work was hugely influenced by being immersed in commercial technology at the Swiss patent office.  Quantum physics was an accidental discovery made by German scientists who were trying to design more accurate instruments to measure very small weights.  The Manhattan Project wasn’t about meeting commercial needs, but as research goes, it’s about as applied as it gets.  Etc., etc.

The point here is that there are parts of commercial science that are up banging against the frontiers of the unknown just as much as university science is: just think of what was discovered at Bell Labs, or what Craig Ventner has accomplished.  It’s where the rubber hits the road: where the most advanced academic science gets put into practice and tested in real-world conditions.  Under commercial pressure, commercial science looks for every little advantage when learning how to cure disease, design better buildings, and develop new technology.

Even Vannevar Bush didn’t believe “pure” research happened in a vacuum.  Indeed, the justification for “pure” research is always that someone, somewhere, will find an application for it.  If you don’t have an inkling of where your “pure” research findings might actually be applied someday, you probably aren’t conducting your “pure” research in a way that’s very effective, because you’re not asking the right questions.

And this is the real reason universities need to engage with industry: it’s where the best questions are.  And you’re not going to get top-notch research without top-notch questions.

October 17

Innovation Literature Fail

So, I’ve been reading Mariana Mazzucato’s, The Entrepreneurial State.  It’s brilliant and irritating, in equal measures.  Brilliant because of the way it skewers certain free-market riffs about the role of risk and entrepreneurialism in the innovation process, and irritating because it’s maddeningly cavalier about applying business terms to government processes (in particular, the term “risk”, which Mazzucato doesn’t seem to understand means something entirely different in government, if losses can be made whole through taxation).

Anyways, one thing that occurred to me while reading was just how America-specific much of the literature on innovation is.  Take the Defence Advanced Research Projects Agency (DARPA).  In innovation policy circles it’s generally considered a wicked-cool way of organizing Big Science: it’s project-based, it brings teams together from both academia and business, and it has substantial independence.  And, of course, the basic research has produced things like GPS and the Internet (still the core anecdotes used to back the “government-should-be-involved-in-research” argument). 

Brilliant, right?  So why doesn’t everyone have a DARPA?  Why doesn’t Canada?

The answer is that DARPA wouldn’t make any sense here.  Our government agencies don’t have enough of the “big problems” that DARPA is designed to solve – or, at least, that could be solved at a price we can afford.  And frankly, we don’t have enough private-sector research scientists to make headway into these kinds of projects, anyway.

More broadly, the American system of funding science works because of a particular combination of factors: the problems needing to be solved, the presence of major private sector research efforts, a particular type of venture capital industry, and scale.  Canada – like most countries in the world – would, at most, get part-marks on any of those four criteria.  So why do we think that policies based on American examples work for us?

Take questions of “applied” vs. “basic” science.  Maybe the classic Vannevar Bush formulation of, “government funds universities to do basic research, and companies do the applied stuff” only makes sense in the US context.  Maybe without the VC culture, or the private sector research culture, the idea that government should only be playing in the “basic” side of the continuum doesn’t make any sense. Maybe countries who aren’t quite at the technological frontier don’t get as much bang for their buck in basic research as America does.

This is just speculation on my part, of course.  But I’m tired of the innovation literature assuming that US-inspired solutions will work here.  Just for once, I’d like to see some literature and policy prescriptions based on what works in Korea, the Netherlands, and Scandinavia.  There’s probably a whole other set of policy lessons to be learned, if only we looked for them in the right places.