Saturday, 27 December 2008

Kealey on scientific motivations and incentives


Invisible colleges, private patronage and commercial profits versus public goods, government funding and ‘crowding-out’: Terence Kealey on the motivations and incentives driving science

Bruce G. Charlton

Medical Hypotheses. 2009; Volume 72, Pages 111-115



What kind of a thing is science and how does it work? [Kealey T. Sex, science and profits: In a recent book (Sex, science and profits: how people evolved to make money. London: William Heinemann; 2008) (p. 455)] Terence Kealey argues persuasively that the motivations driving science are widely misunderstood. Science is often assumed to be useful to the public but an economic loser for the scientist and his or her paymasters – in other words, science is supposed to be a ‘public good’. The public good argument is used to support large-scale government funding of science, on the basis that if government does not fund science it will not be funded adequately. But Kealey argues that most science is profitable to commercial organizations, and other types of worthwhile science will be supported by private patronage. Yet excessive government funding tends to ‘crowd-out’ potential private sources of funding – both by replacing and by deterring private investment. And scientists are not primarily motivated by money, but instead by striving for status within the ‘invisible college’ of active researchers in their field. Kealey’s take-home message is that overall and in the long-term, science neither requires nor benefits from government funding. Scientific research would be better-served by private funding from commercial organizations that are seeking profit, combined with patronage from charities and foundations that regard science as intrinsically valuable.


What kind of a thing is science and how does it work? In a recent book [1] – tightly packed with insights, evidence and jokes – Terence Kealey argues persuasively that the motivations driving science are widely misunderstood. Kealey suggests that the major motivation for scientists is to attain status within the ‘invisible college’ of active researchers in their field. So powerful is this motivation that scientists routinely subsidise their work from their own pockets.

Science is often assumed to be useful to the public but an economic loser for the scientist and his or her paymasters – in other words, science is supposed to be a ‘public good’. The public good argument is used to support large-scale government funding of science, on the basis that if government does not fund science it will not be funded adequately. But Kealey argues that most science is profitable to commercial organizations, and other worthwhile types of science will be supported by private patronage.

Yet excessive government funding tends to ‘crowd-out’ potential private sources of funding – both by replacing and by deterring private investment. Kealey’s take-home message is that science neither requires, nor benefits from, government funding; but that over the long-term science would be better-served by private funding by commercial organizations that are seeking profit combined with patronage by charities and foundations that regard science as intrinsically valuable.
Status motivates scientists

Although, like almost everyone else, scientists need income (and would prefer to have more of it), scientists are not ‘in it for the money’. Instead, scientists are ‘in it for the status’ – specifically the esteem of the peer group of scientists actively working in the same field.

The ‘sex’ element in the book’s title explains this, since it refers to sexual selection [2] and [3]. Sexual selection provides the ultimate evolutionary explanation as to the human motivation to seek status, because status is attractive – especially male status is attractive to females [4]. Science can be seen as one of many competitive social systems which harnesses and puts-to-good-use the human craving for prestige within a social group.

If scientists are primarily motivated by the desire for status rather than money, then this predicts that they will risk or sacrifice money to achieve scientific status. And this can often be observed – much scientific activity (e.g. computer, book and stationary purchases; journal subscriptions; conference attendance; media appearances; advice and consultancy; even postgraduate training) is funded to some extent from scientists’ own pockets or done in their private time.

In one of numerous witty and telling asides, Kealey comments that in economic terms virtually all science publishing is a type of ‘vanity publishing’ (a term usually reserved for the self-publication of slim volumes of verse) – since scientific papers are published at best with no monetary reward, and more often at significant monetary loss. Authors are prepared to pay when this helps to disseminate their research and scholarship: examples of ‘pay-to-publish’ journals are legion and include Proceedings of the National Academy of Sciences (PNAS), PLoS Medicine, and indeed Medical Hypotheses.

Tacit knowledge and the invisible college

The group within which a scientist seeks status is termed an ‘invisible college’ [5] Wikipedia. Invisible College. (accessed 12.09.08).[5] – ‘invisible’ in the sense that it is characterized by the exchange of information among dispersed peers rather than by co-residence in a specific physical building, and a ‘college’ in the sense that it is exclusive. Nowadays an invisible college is often international and may be sustained by electronic communications rather than face-to-face interactions. Nonetheless, Kealey emphasizes that personal relationships remains very important in science, as in many other areas of modern life, as evidenced by the phenomenon of ‘clustering’, in which rival niche businesses tend to locate near to each other. Clustering facilitates mutual evaluation of personal qualities and the build-up of networks of trust that allow information exchange. Conferences perform a similar function in science.

But the invisible college is not about giving-away information to all-and-sundry; the invisible college is specifically about trading information for mutual benefit. And the trading of knowledge is a matter of enlightened self-interest – in the long-term a group who share knowledge will out-perform those who do not share. However, the fact that an invisible college is based on trade means that participants must have something to trade which is valued by the other members of the college. Those with no useful knowledge to trade are excluded from the college.

One of the types of information which is traded in science is ‘tacit knowledge’ (the term comes from the philosopher and scientist Michael Polanyi [6]). In science tacit knowledge includes a great deal of unpublished, and probably un-publishable, knowledge about the skills and minutiae of doing science – the kind of thing that can only be learnt from direct personal apprenticeship – for example during a Ph.D. It is tacit knowledge which is learned by personal conversation, or lab visits, and by working alongside experts – for example in discovering and dealing-with raw data.

So, there is a substantial cost attached to membership of an invisible college. The cost involves attaining sufficient expertise to participate in college communications; and on top of that there is the effort required to generate new knowledge or skills which can be traded: i.e. the possession of knowledge or skills for which other scientists are willing to trade their own tacit knowledge.

The trading of knowledge depends on trust – the greater the level of trust the more extended and complex can be the networks of mutually-beneficial information exchange. But if trust is betrayed so that valid and useful information flows only one way, then the benefits of trade become merely one-sided and ultimately trade breaks-down [7]. And trust must be earned, which is another reason for the exclusivity of invisible colleges. Those who have not yet proved themselves trustworthy, or who have behaved in an untrustworthy manner, will not get access to tacit knowledge.

To what extent is copying a problem?

It is easy enough to acknowledge the workings of invisible colleges in non-commercial research, but Kealey emphasizes how exactly the same mechanisms support scientific research in the commercial, profit-seeking, sector. He argues that, far from being secretive, competitors in the private sector share their scientific knowledge – for the same self-interested reasons as in the non-profit sector: that those who share knowledge out-perform those who do not. However, private firms are selective concerning whom their knowledge is shared-with, and there must be trust in reciprocity. And most relevant and useful knowledge is either highly technical or tacit, and therefore, either incomprehensible or inaccessible to those outside the invisible college.

It is generally assumed that the major problem for commercial science is the danger of copying, whereby a commercial rival takes the benefits of innovation without paying the cost of development. Of course, in one sense, copying is the basis of human progress, the mechanism by which improvements spread from their originator. Unless there was copying there would be no general benefit. So, the real problem is attaining a balance between the public benefits of copying and the need for the originator to have an incentive to innovate. Copying is good, but there should also be the probability of a sufficient reward for the originator.

The question is what form this reward should take. Since scientists are primarily seeking status within the invisible college, they often welcome copying – so long as the innovator gets personal credit for having the original idea. In other words, since scientists are primarily motivated by status rather than money, it is logical that their rewards should primarily be in the currency of prestige rather than cash. From this ‘pure’ science perspective, plagiarism – i.e. copying without giving credit to the source – is a problem; but copying with credit is actually in itself a reward because it confers status.

But Kealey emphasizes that even in commercial business situations copying is not such a big problem as is sometimes imagined. The reason is that copying is usually costly: very costly. Copying takes both time and considerable money. Furthermore copying involves other scarce and valuable resources too, because a firm which is set-up to copy its rivals must support a high level of scientific expertise in order that they can detect and understand potentially-useful innovations as well as doing the actual copying and production.

In other words, although apparently ‘free’ and seldom kept secret, scientific knowledge is in fact comprehensible and usable by only a small and highly-trained elite group. Relevant tacit knowledge (without which explicit knowledge can seldom be exploited) is a kind of ‘secret ingredient’ that is extremely costly to acquire because it is usually only traded for equally valuable tacit knowledge – and this keeps it within the exclusive confines of the trust-based invisible college.

Kealey suggests that it is the intrinsic difficulty of copying, rather than ‘intellectual property rights’ such as patents, which are the significant factors in protecting the monopoly of innovators, allowing them to profit from their innovations and thereby providing the motivation for innovation.

If so, then the major commercial advantage of innovation is to gain the ‘first mover advantage’ in monopolizing a market niche: the temporary monopoly situation that exists before rivals are able to mobilize production, distribution and marketing of their copies. As a consequence Kealey suggests that patents are usually a bad thing all-round, both for the public and the patenter.

(There are exceptions in rare and specific circumstances where there is a long and expensive investment to generate a product which is relatively quick and cheap to copy – such as the pharmaceutical industry.)

An example of patents harming the public is James Watt’s steam engine patent, which held-back progress in the field for several decades. Furthermore patents can also distract innovators from the primary requirement rapidly to exploit their first mover advantage, and to continue seeking further innovations. An example of patents harming the patenter is the Wright brothers Wilbur and Orville, who invented the first functional aeroplane; but squandered prodigious time, energy and money in trying (and usually failing) to defend their patents.

If this is correct, and the argument seems plausible, then there needs to be a radical reappraisal of the current trend progressively to extend the scope and duration of intellectual property rights – especially copyright. If the aim is to benefit the public, probably we ought to be experimenting with narrowing the scope of patenting, and progressive shortening of the allowed duration of copyright and patents in most intellectual fields.

Government funding and ‘crowding-out’

By describing a range of powerful economic and personal motivations to do science, Kealey is arguing the falsity of the common conceptualization of science as a public good requiring government funding. His idea is that ‘self-interest’ of various types (individual and institutional) would ensure that enough science of the right kind was funded without need for government intervention.

Furthermore Kealey suggests that increasing government funding of science may not lead to greater funding overall, because state support ‘crowds out’ and deters private funding, leading to excess dependence on government funding and the damaging politicization of science.

It is extremely important to be aware of, and to recognize, the possibility of crowding-out of private by state investment in science. Because those of us who live in the developed world are used to prodigious and rising levels of government funding for science – doubling in real terms about every 15 years for science overall [1] and with even more rapid growth (doubling about every decade) for medical science [7]. Kealey provides evidence that such expansion of government funding crowds out private funding in science. Crowding-out means that in practice (although not necessarily by intention) private funding usually declines as government funding increases until science becomes de facto almost a public monopoly.

Of course, most people assume that government funding will cause crowding-in to science – in other words they hope that the more money which government spends, the more private money will be attracted to the field. Kealey’s example of crowding-in occurs in road transportation when the government builts a new road with free usage, and private individuals then buy more cars to use the improved facility.

But the evidence suggests that increasing government funding of science usually crowds out private funding, leading – for instance – to reduced corporate investment in research and development (and Kealey emphasizes that R&D really is an investment, making a significant contribution to organizational profitability). The more that government funding increases, the more that private funding declines. Possible reasons for crowding-out the private sector include state money being used to substitute for private research funding, and the general reduction in competition between rival firms which is often explicitly encouraged by government policy.

Another factor in the phenomenon of government funding crowding-out private funding is, I suspect, probably related to science being a status-oriented activity. Where government provides the lion’s share of funding then government will take the lion’s share of status. Because status is a zero-sum game, if one party gets a larger share of prestige then this leaves less prestige for everyone else. There is little incentive for private patrons to engage in funding science if government then claims most of the credit for the achievements of science – so discriminating patrons will look elsewhere to gain prestige for allocating their support.

Creative destruction

Governments often believe that competition is harmful, in science as elsewhere, and they instead try to encourage cooperation. But Kealey agrees with Joseph Schumpeter’s general economic schema of capitalism as powered-by ‘creative destruction’ – an interpretation which emphasizes the value of competition in driving innovation, and innovation in promoting economic growth [8].

According to Schumpeter, the main driving force of capitalism is that companies compete to gain then retain monopoly. Entrepreneurs continuously pursue innovation and where this is successful the price of failure paid by commercial rivals may be extinction (i.e. destruction). Failure is intrinsic to the dynamic of success [9]. Incumbent monopolists are forced to seek innovation by the credible threat of rival entrepreneurs ‘snapping at their heels’ by generating innovations intended to take away their markets.

This destruction of commercial firms is termed ‘creative’ because in the long-term it allows better allocation of resources – innovation can only thrive at the expense of discarding the obsolete. A few hundred years ago some 90% of the UK population were engaged in growing food, now the proportion working in agriculture is less than 1%, and the other 89% of the ex-farm-labouring population have (forcibly) liberated to perform a multitude of other economic activities.

The destruction of agriculture as the dominant mode of employment was therefore, creative in the sense that it enabled a massive reallocation of manpower with an increase in productivity (economic output per person), and therefore, creative destruction of agriculture promoted economic growth. Failure in one area was necessary to success in another. Specific short-term suffering (e.g. reduced income among newly unemployed farm labourers) was intrinsic to the general long-term benefits (e.g. increased income spread among everyone, including the grandchildren of the unemployed farm labourers).

Creative destruction is, of course, also a feature of science: new theories displace the old and the destruction of old ideas is necessary in order to liberate resources to invest in new science. Karl Popper’s philosophy of science could, indeed, be regarded as an equivalent process to Schumpeter’s creative destruction in the economy [10]. Example are legion: the decline of astrology entailed by the creation of astronomy, the extinction of alchemy and the reallocation of effort into chemistry, the replacement of classical physics by relativity and quantum theory [11].

The progress of science generally and in the long-term is attainable only at the cost of specific and short-term destruction – e.g. of the reputations and livelihoods of scientists working in fields which have been superseded.

Political corruption of science by excessive government influence

Because expanding government support of science usually results in crowding-out of private support, the state tends to become a monopoly supplier of science. Government funding agencies then shape the direction of scientific research, and government thereby gains an unhealthy degree of control. Essentially, the systems of government and science fuse to form a single hybrid system of politicized science – i.e. science that promotes government interests.

At the end-point of decades of crowding-out, all kinds of pathologies of science have now emerged, although they are often taken for granted and hardly noticed. For example, in the UK, government-funded science is often regarded (especially by other government agencies) as having intrinsically higher prestige than privately-funded science. This principle has been formalized in some UK state-administered university research evaluation mechanisms (RAE – [12]) where extra credit was given to funding from government agencies (in my opinion credit should only have been given for scientific output, and not for input measures such as funding [13]).

Another almost inevitable effect of government funding is inefficiency [7]. As science becomes more like a monopolistic ‘nationalized industry’ there are the usual problems of excessive yet still-growing bureaucracy. Many economists recognize as a general principle that government provision is only about half as efficient (in terms of outputs per input, and controlling for quality) as the private sector [14].

Science funding may be used to pursue political, not scientific, goals. For example government may shift funding towards what are perceived as vote-winning causes that are likely to attract votes such as the ‘war on’ cancer, or special attention on AIDS, breast cancer or other diseases. The special government attention may or may not be scientifically or medically justifiable – but the reason for extra funding is primarily political. Or political parties may deploy large-scale research facilities to specific geographic regions as a reward for political support. Or parties may ‘buy’ the support of scientists by (in effect) bribing them with extra grants and salaries, or blackmailing scientists with the (credible) threat of bribe-withdrawal if the opposition party gets power.

And having crowded-out significant private funding from science, government may then shape the emphasis of science strategy so that they build-up science and scientific perspectives which bolster their political views while simultaneously starving-out science that is hostile to the governing ethos. The most notorious example of this was Stalin’s support of ‘Lysenkoism’ in the mid-twentieth century USSR and the persecution of biologists who adhered to Darwin’s theory of natural selection [15]. A similar phenomenon (usually minus the violence) can be seen in the way that IQ research has been marginalized by government policy in the US and UK [16].

Large-scale and long-term government funding of science creates economic and psychological dependence on the state. It is this pervasive dependency culture which makes Kealey’s message so alarming and unacceptable to most scientists.

A vision of diverse private science funding

Kealey’s ideal is a world in which government funding of science has dwindled to insignificant levels, and science is mainly funded by a mixture of private commerce and private patronage.

Commercial organizations would be motivated to do science by the bottom-line of profit; because pursuing continued innovation and supporting science is the only way they can have access to the invisible college of information (including tacit knowledge) upon which their continued competitiveness depends.

In the absence of crowding-out by government funding, substantial further science patronage would come from private research foundations and charities of many types, who would fund diverse branches of science simply because they value that science and like to support it – just as foundations and charities support the arts, sports, religions and numerous other good causes. This altruistic motivation is helped by the fact that supporting science is, in general, a prestigious thing to do.

Scientists are motivated to work mainly by the desire for enhanced status within their invisible college – which ensures competition; they are also motivated to trade their knowledge (i.e. cooperate) by the advantages this brings to their competitive pursuit of scientific knowledge. For these reasons scientists will spontaneously self-organize into exclusive and elite invisible colleges or communities of reciprocal trust for mutual benefit.

So, in a world of negligible state funding of science; competition and cooperation would work together at both individual and organizational levels to create a dynamic and self-regulating system orientated towards increased efficiency and growth of knowledge by means of creative destruction.

Modern science may be over-funded overall, and the pattern of funding sub-optimal

Kealey’s book provokes me to further speculation.

In Kealy’s world of mostly-private funding of science it seems possible, or likely, that the total quantity of private funding for science would differ significantly from the current level of state funding.

In other words, government might – according to circumstances – either be over-funding or under-funding science overall. Furthermore, it is likely that the pattern of state funding of science (i.e. the distribution of funding between specialties) would be different from the spontaneous outcome of a private system of funding.

Indeed, lacking a market, hence lacking ‘price’ signals, it may not be possible to know – even in principle – whether a system such as science was optimally-funded [17] and [18]. From a ‘systems theory’ perspective, it could be argued that a private (‘market’) system of funding would be able to reach a more efficient volume and distribution of funding for science than could centralized decision-making [19].

If science was under-funded by government, the spontaneous tendency would be for private funding to move-in to fill the gaps where needed. But science could be chronically under-funded if private companies and patrons were legislatively-prevented or financially-deterred from spending money on science. Or science may be blocked by political, religious or ethical considerations – especially in totalitarian societies. And even in liberal democracies, scientific research may become so hemmed-around with regulations, restrictions or potential legal pitfalls as to prevent, or at least deter, private spending.

Alternatively, or at the same time as there is chronic under-funding in some areas, other areas of science could be chronically over-funded by the government, especially if government spending was driven by political rather than scientific objectives, as described above. Indeed, my guess is that medical science is currently over-funded in the US, UK and Western Europe [7], because medical research is used as a way of government gathering voter support by spending public money to show that it ‘cares’.

And overall over-funding of scientific research may tend to be self-sustaining if it creates a bigger and more powerful interest group to lobby in favour of continued over-funding. This interest group would extend beyond scientific researchers to embrace universities and the science communications media.

If science was indeed over-funded by government, then there would be a decline in funding under private provision. Add to this the inevitable reallocation of funding between different branches of science which would result for any change in the funding mechanism, and these threats will mobilize interest groups to resist change – even when such change would be beneficial overall and in the long-term.

Of course, it is very likely that a private funding system would open-up many new areas of science that are currently starved of support; however, these areas of science (precisely because they have been starved!) currently lack any incumbents who might lobby for such policy changes.

So normal ‘interest politics’, politics as usual, will usually tend to prevent radical change to the scientific research funding allocation procedures – except when change benefits the most powerful incumbents.
Five (bad) reasons why Kealey’s analysis will be ignored (especially if correct)

I believe that Terence Kealey is generally correct in his analysis and arguments. However, I am pessimistic about the prospects for these ideas getting serious consideration.

There are at least five powerful aspects of mainstream scientific thinking which make it probable that the arguments will be ignored:

1. There is now a long-established dependency culture of science. Scientists find it hard even to imagine a world without quasi-monopolistic government funding of science.

2. This dependency culture of science has for decades been attracting into the profession a risk-averse and uncompetitive type of scientist who prefers the mediocre conditions and job-security of working under bureaucratic supervision to the freer and more meritocratic – but harsher and less predictable – world of ‘creative destruction’ associated with markets and private funding.

3. There is, I believe a covert suspicion among many scientists that science in the US, UK and Western Europe may be over-funded, and therefore, that any shift to a more efficient and effective mode of supporting science privately would lead (overall) to a significant cull of scientists and ‘down-sizing’ of scientific research organizations.

4. The current scientific leadership has evolved a cozy, collusive and mutually-beneficial relationship with government; and this would be threatened if there was a significant diminution of state support for science. For instance, in the UK the Royal Society gets more than two thirds of its income direct from the UK government. Senior personnel migrate back and forth between science and the civil service. A major change towards private funding would bring the potential for a major loss of power and prestige for many powerful figures in science administration.

5. Combining all the above points – if science funding was reformed, things would get worse before they got better. Creative destruction may be the way that the world improves; but, since destruction comes before creation, blame for early damage is almost inevitably allocated with greater surety than credit for later benefits. Those responsible for policies that trigger short-term harm to science and scientists will therefore, be subject to the certainty of immediate vilification and the possibility of an enduring negative reputation. These are major reasons for bad government everywhere and at all times.

In sum, sensible reform of science funding away from quasi-monopolistic government support and towards a multitude of private sources will probably have to wait until, eventually, some desperate crisis will force the appropriate action.


[1] T. Kealey, Sex, science and profits: how people evolved to make money, William Heinemann, London (2008) p. 455.

[2] H. Cronin, The ant and the peacock: altruism and sexual selection from Darwin to today, Cambridge University Press, New York, UK (1993).

[3] G. Miller, The mating mind: how sexual choice shaped the evolution of human nature, Heinemann, Oxford, UK (2000).

[4] D.M. Buss, The evolution of desire: strategies of human mating, Basic Books, New York (1994).

[5] Wikipedia. Invisible College. (accessed 12.09.08).

[6] R. Allen, Polanyi, Claridge Press, London (1990).

[7] B.G. Charlton, Boom or bubble? Is medical research thriving or about to crash?, Med Hypotheses 66 (2006), pp. 1–2.

[8] J.A. Schumpeter, Capitalism, socialism and democracy, Allen and Unwin, London (1954).

[9] P. Ormerod, Why most things fail: evolution, extinction and economics, Wiley, Hoboken, NJ, USA (2007).

[10] B. Magee, Popper, Collins, London (1973).

[11] J. Bronowski, The ascent of man, BBC, London (1983).

[12] Wikipedia. Research Assessment Exercise. (accessed 12.09.08).

[13] B.G. Charlton and P. Andras, Evaluating universities using simple scientometric research-output metrics: total citation counts per university for a retrospective seven-year rolling sample, Science and Public Policy 34 (2007), pp. 555–563.

[14] M. Friedman, Capitalism and freedom, University of Chicago Press, Chicago (1963).

[15] D. Joravsky, The Lysenko affair, Harvard University Press, Cambridge, MA, USA (1970).

[16] Charlton BG. Pioneering studies of IQ by G.H. Thomson and J.F. Duff – an example of established knowledge subsequently ‘hidden in plain sight’. Medical Hypotheses 2008;71:625-8.

[17] T. Sowell, Knowledge and decisions, Basic Books, New York (1980).

[18] F.A. von Hayek, The fatal conceit, Routledge, London (1988).

[19] B. Charlton and P. Andras, The modernization imperative, Imprint Academic, Exeter, UK (2003).