It is a mistake to eliminate government science. Part I

Figure 1. From the Department of Fisheries and Oceans website, an image of one of the Experimental Lakes in northern Ontario.

UPDATE:  If you are interested in efforts to save the Experimental Lakes Research Station please visit this site.  Part II of this post is here.

There has been coverage lately about the elimination of many government supported research programs in Canada.  The Experimental Lakes Research Area (coverage here, here, here and here)  is the latest in a string of cuts aimed at reducing the federal deficit.  Much of the decision rests on the idea that scientific work can be outsourced from government to Universities (as it already is through funding arrangements between researchers and NSERC) or to private contractors.  Much of this discussion will focus on ecology and ecological research since that is my field of interest.  I will expand a bit and discuss the state of affairs in biomedical research, a field where there is considerable privatization of research in Part II (link to come).

Some research simply can’t be done on the basis of contracts.  The Experimental Lakes Research Station that I linked to earlier is a good example.  To do research on timescales that are ecologically meaningful you need to maintain overhead and some sort of central structure, and you need to provide long-term support for the data generated so that it can remain accessible.  You can’t contract that out to a university (through individual researchers) because professors change jobs or retire.  You need an administrative structure in place with long term funding guarantees.  Given that, when you remove existing infrastructure (by cutting government departments and research programs) you lose the central capacity to maintain existing data and support current research programs, except in a limited form. Ultimately that data would be lost without support from the government (through organizational data rescue plans), at a cost that is likely higher than the ongoing support necessary to prevent loss.

Some Ph.D students (now post-docs) working hard at their other job. Rocking.

I don’t disagree that there is a problem with federal finances that needs to be addressed.  Unfortunately, contracting and outsourcing as a solution will not work unless government converts departmental research money to some other form (such as independent contracts). If outsourcing is the goal they’d put money into postdoc scholarships (my plug follows!), we work incredibly cheap and have an enormous vested interest in generating ‘product’ since our careers depend on it, but they’ve cut the NSERC postdoc funding by more than half (from 286 in 2010 to 133 in 2011).  They could fund PhD students and MSc students but they’ve cut those programs too (see previous links, from 1219 and 1301 respectively in 2010 to 876 and 798 in 2011), and now the basic fundingto maintain and purchase research infrastructure for labs and special collections is being cut too.  The cuts to research infrastructure are particularly key because they make it increasingly difficult for established profs to maintain their labs and long term projects.

Ultimately, if they’re going to outsource research they need to maintain intellectual capacity to judge the merit of the outsources research products, but it’s just not there.  The government has simultaneously fired researchers in a number of federal departments and cut basic research funding at universities, so federal researchers are essentially lost to the ideas marketplace, unless private industry can facilitate the transition.  I argue that it cannot under the current system.

I can’t pay to research the effects of fertilizer treatments on crop yields and ecosystem functioning all by myself, but that research directly impacts me! (click for source)

For private industry to enter there need to be funding guarantees in place to support the kinds of basic research that government used to provide. This is particularly true for projects that might have a ‘distributed’ set of end users (citizens, farmers, municipalities), who, individually, would be unable to support the kinds of research that would provide them with valuable information about water quality, climate variability, forecasting and other ‘large-scale’ topics.  I’m very much a pragmatist when it comes to the value of ecological research (much to the consternation of some), while I believe there may be inherent value to maintaining and understanding natural systems, there is a fundamental need to define that value so that you can provide rationale for funding, especially when you invoke the private sector.  It becomes a trade off, either the government funds basic research under a loose assumption of inherent value, or we need to explicitly define the value of ecological systems and make end users pay for the research to maintain the ecological systems.  If we chose the second option  then you need to undertake a valuation exercise before you switch to the price enterprise system, otherwise you lose the value of your investment (e.g.: crown property and the ecological services provided therein).

In some cases ecological valuation isn’t too hard.  You can make a company pay for an environmental assessment when development impacts only a single site.  Unfortunately, the ability to define a single end user for privatized research becomes more difficult as you scale projects.  Scaling means you are dealing with esoteric issues (what is the value of a rare species and what may affect its rarity) and you begin to perceive networks of impacted systems where effects across scales might not be additive.  Not only can you not define an end user, but you can’t construct a single study to understand these processes and you need long-term data from multiple sources.  All of this needs coordinated infrastructure that industry cannot provide, that even universities cannot provide.  These resources are and should remain the domain of government.

More to come in Part II.

Music:  California – Rogue Wave