Title: The Economic Laws of Scientific Research
Author: Terence Kealey
Scope: 3 stars
Readability: 3 stars
My personal rating: 4 stars
See more on my book rating system.

If you enjoy this summary, please support the author by buying the book.

Topic of Book

Kealey examines whether government-funded research promotes economic growth.

If you would like to learn more about role of scientific research in promoting progress, read my book From Poverty to Progress: How Humans Invented Progress, and How We Can Keep It Going.

Key Take-aways

• Since Francis Bacon, the dominant paradigm for research has been that government funding of basic science will lead to technological innovation which will in turn leads to economic growth.
• Alternatively, Adam Smith believed that new technology comes from industrial improvements to existing technology. Academic science played only a small role.
• In Britain and the USA during the Industrial Revolution, scientists were self-funded hobbyists or were funded by industry or private organizations.
• Then during WWI, WWII and the Cold War, the governments of those nations increasingly funded science for military reasons.
• Even after the end of the wars, the new funding continued because the scientists want the funding.
• As nations increase in wealth, they devote greater resources to science regardless of government policy. Poorer nations do not need to fund science, because they can copy the rich nations.
• Government funding of science tends to crowd out private funding because of the finite number of scientists. In addition, it changes their purpose toward political or military ends.
• Basic science, applied science and technology are relatively isolated from each other. Progress within the technology field comes from recombination of existing technologies, not breakthroughs in science. Same for the other fields.
• The greatest cross-fertilization among basic science, applied science and technology comes from the opposite direction: new technologies make scientific research much productive.
• The biggest technological breakthroughs coming from science come not from the original researchers who made the breakthrough. They come from those who read their writing, understand the technological implications, and then create a new technology based upon those principles. This is the second-mover advantage.
• The problem is that only scientists actively working within the field can understand the work of other scientists. That is why private industry pays for scientists: not for what they do, but for what they read.

Important Quotes from Book

During the twentieth century the governments of all advanced countries started to fund research… This book was written to make an audit of these different policies. Should governments fund science? If so, by how much? If not, why not? The debate is an old one, and it goes back to two Britons, Francis Bacon and Adam Smith.

Bacon … invented a new intellectual model. Where medieval philosophers had advocated deduction, Bacon honoured induction as the finest of scientific achievements: Bacon believed that a researcher, on discovering new facts, could thus induce new laws of nature. Bacon went further still – he invented the concept of progress, or progression as he called it.

It was Bacon, therefore, who first proposed the ‘linear’ model of technological advance:

• Government-funded academic research leads to:
• Pure Science, which leads to:
• Applied Science or technology, which leads to:
• Economic growth

Smith, however, disagreed with Bacon over the source of technology. Bacon supposed it would flow from academic science, Smith observed that it arose from within industry itself.

Smith enumerated the three major sources of new technology. Pride of place, again, went to the factories:.. Almost as important as a source of new technology, in Smith’s judgment, were the factories that supplied other factories.. Only third, and least important, was the input that flowed from academic sciences.

Where Bacon believed that it would flow from academic science, Smith maintained that it largely derived from the industrial development of pre-existing technology. Indeed, Smith went even further – he actually reversed the intellectual flow – he believed that most advances in science were not made by academics at all, but by industrialists or by others working outside academia.

In summary, Smith did not believe that applied science flowed very much from pure science; indeed, he believed the opposite was as likely to be true. Moreover, he believed that applied science or technology sprang from the market place, spawned by individuals or companies competing for profits. In as much as advances in technology did emerge from pure science, Smith did not believe that that justified government spending. Smith feared the economic consequences of the increased taxation that would entail, he distrusted any measure that increased the power of politicians, and he distrusted the capacity of academics to work without immediate goads. He believed that private sources (students’ fees, endowments and consultancies) would provide enough funding for university science to meet the needs of the economy.

We can test their models against the historical experience.

The Industrial Revolution did not represent the application of science to technology, it represented the development of pre-existing technology by hands-on technologists.

The industrial revolution was created by men looking for solutions to very particular problems – men who had the economic freedom and the economic incentive to invest their time and resources in experimentation and development.

We can see, therefore, why Britain – not France – pioneered the Industrial Revolution. Intellectuals on both sides of the channel championed Bacon and state funding, but intellectuals had no power in free enterprise Britain, where only the market ruled, so Britain invested its money appropriately, in technology. The French listened to their intellectuals, and dissipated their money on science.

Laissez faire Britain created a flourishing science – witness such names as Davy, Faraday, Maxwell, Dalton, Kelvin, Darwin, Huxley, Lyell, Cavendish and Rosse. How was this funded? There were at least five separate sources of funding for science under the free market in Britain: (i) hobby science; (ii) industrial science; (iii) university science that had been endowed by industry; (iv) university science that had been endowed privately; and (v) university science that was funded out of fees or commissions.

Britain funded pure science because, as the Industrial Revolution developed, industrialists increasingly perceived that they were exhausting technological development, and they needed to explore science more profoundly. Since the Government refused to do it, and since the industrialists were not anxious to pay any extra taxes to enable the Government to do it, they invested their own money in science.

We can construct, therefore, from these studies, a general economic law: the richer the country, the greater the percentage of its wealth it spends on R&D, the more science it does, and the better that science is.

Poor countries enrich themselves by copying. Poor countries are poor because their technology is inferior and their work force inefficient, but they do not need to engage in R&D to correct the deficiencies; all they have to do is to copy the rich countries.

Poor countries, therefore, enrich themselves by copying or by importing expertise, they do not need to do their own R&D. Rich countries, however, have no one to copy. If they are to get richer, then they can only do so by innovation so they invest in R&D. But this creates a twist: copying is much easier than innovation (for obvious reasons) so poor countries can grow faster than rich ones. Much economic history since 1870, therefore, becomes easy to understand. Certain countries, particularly Britain, Australia, Belgium and Holland had, before 1870, through their embrace of capitalism, grown rich. That enrichment had happened because capitalism itself had generated the money for R&D. Other countries, on observing the success of the four lead countries, started actively to copy them. Because copying is easier than innovation, these other countries enjoyed astonishing rates of growth; because innovation is difficult, the lead countries have always grown slowly.

Until the 1890s America, being poorer than Britain, was also technologically inferior. It was enough for American engineers to copy British practice. But after 1890 Americans, being in the economic lead, encountered novel technological problems. These could not be resolved just by copying- engineers needed to be good scientists. The need for academic training suddenly became pre-eminent, and thereafter the colleges took over the training of the entire profession. Thus did laissez faire direct education appropriately, because after 1890 employers would only hire college-trained engineers, who therefore commanded higher salaries to commute their college fees (just as do American doctors today).

1870 has failed to be very interesting. The pioneering industrialised countries have continued to grow economically, but relatively slowly, because they have largely grown through R&D. Those poorer countries that have caught up have done so by copying the technology of the lead countries, which has blessed them with high rates of initial growth. But on enriching themselves, they too have had to settle for lower rates of economic growth and for increasing their investment in R&D.

We shall largely concentrate on two countries, the USA and the UK, because they share a common and pivotal history. Each was, in tum, the lead country economically, becoming so while pursuing laissez faire policies for science. Each, however, is now scientifically dirigiste.

The manner of the foundation of the National Academy of Sciences was characteristic. There will always be scientists who believe that science should be advanced with government funds, governments will resist the pressure but, time again, war persuades them to create scientific institutions. Once created, institutions are hard to destroy; they invent justifications for their continued existence, and they mobilise political support. One feature of scientific institutions is their intellectual flexibility. In war they emphasise their potential for creating weapons of death, in peace they explain how they can create wealth.

The parallels between the scientific responses to the Great War and the Civil War were extraordinary. New agencies were created (the Advisory Committee for Aeronautics, for example, which spent over $1 million of federal funds between 1915 and 1919) but, uncannily, old agencies were re-created under new names.

If there is one constant in science, it is that at any one time there is at least one powerful person, waiting in the wings, just itching to impose a centrally planned, federally funded science policy on the people of America. Most of these people die ·unsung, but if they are fortunate, and a war just happens to coincide with their career peaks, then they soar into apotheosis. The Civil War enabled Bache to create the National Academy of Sciences, the Great War enabled Hale to create the National Research Council, and the Second World War provided Vannevar Bush with his opportunity.

Let us review the consequences of the federal government’s post-1940 research policies. In 1940 (see Table 8.1) R&D in the USA was dominated by the private sector, which spent $265 million out of the total of $346.1 million. The government only spent $81.1 million, and this concentrated on two sectors, defence and agriculture (successive administrations having adopted agricultural R&D as a legitimate government expenditure to buy the farmers’ votes). The bulk of civil R&D was funded by industry ($234 million in 1940) and basic science, which represented about 10 per cent of the entire civil R&D budget, was dominated by the private sector.

The story can be simply told. In 1940, private industry dominated this sector, but after 1940 the federal government moved in massively, largely to fund defence R&D but also to fund space, health, nuclear research and to maintain its agricultural research (Figures 8.1 and 8.2). But, after the war, private industry also continued to expand its budgets, and private and federal expenditure have marched more or less together; the government concentrating on defense R&D and private industry concentrating on civil R&D.

Between 1820 and 1979 the American economy has grown at a linear rate of about 2 per cent GOP per capita per year. Neither the sudden intervention of government funding for basic science in 1940, nor its soaring peak in 1968, nor its subsequent small decline or revival, seem to have done anything to the economy.

Project Hindsight, whose final report was published by the Office of the Director of Defense Research and Engineering in 1969, concluded that applied science built on applied science, not on pure science. Indeed, the evidence suggested that, if anything, it was the unexpected discoveries made by technologists or engineers that boosted pure science, rather than the other way round.

Here is a central question in Man’s millennial search for a better life: whence does new technology come?

The most comprehensive answer to this question emerges from the study of Edwin Mansfield.Mansfield surveyed 76 major American firms which, collectively, accounted for one-third of all sales in seven crucial manufacturing industries – information processing, electrical equipment, chemicals, instruments, drugs, metals and oil. He discovered that, for the period 1975-85: ‘about 11% of new products and about 9% of new processes could not have been developed, without substantial delay, in the absence of recent academic research’. An earlier survey of American industry by Gellman Associates9 also showed that around 10 per cent of industrial advances were attributable to recent academic research. These findings represent, therefore, a gaping flaw in the linear model: only 10 per cent of new technology emerges from academic research. Ninety per cent of new technology arises from the industrial development of pre-existing technology, not from academic science.

A British study by Langrish et al. spelt this out: reporting on the origins of 84 technical innovations in British industry (so important that they won the Queen’s Award for Industry) they found that ‘although scientific discoveries occasionally lead to new technology, this is rare’ .Generally, ‘technology builds on technology’ and that building of technology occurs within the R&D departments of industry.

Indeed, the implications for the linear model are even grimmer than these statistics show. Mansfield may have found that around 10 per cent of new products or processes arose from the industrial development of academic research, but these innovations tended to be economically marginal: they only accounted for 3 per cent of sales and 1 per cent of the savings that industry made through innovation. These studies, therefore, show that 90 per cent of industrial innovation, and well over 95 per cent of industry’s profits through innovation, arise in-house from the industrial development of pre-existing technology.

The so-called linear model not only requires a fork, it also requires a reverse arrow. Technology can lead as vigorously to advances in basic science as vice versa.

In daily practice, academic science is as self-contained as technology. Just as 90 per cent of new technology arises from old technology so, for most of the time, new academic science builds on old. The two disciplines largely grow separately and I, as a practising biochemist, rarely observe my colleagues reading journals that are not biochemical. The so-called linear model, therefore, requires yet a further modification to reflect that both science and technology are largely self-contained, growing on themselves.

Basic science produces two types of commercial benefit, ‘first-‘ and ‘second-mover advantages’. First-mover advantages (as defined by M. Leiberman and D. Montgomery in their ‘First Mover Advantages’25) are exactly what they are called, they are the advantages that accrue from discovering something first. They are discoveries. The commercial fruits of first mover basic research can be gargantuan, and individual companies can expand to dominate an entire industry through the exploitation of an advance in fundamental knowledge; but basic science is unpredictable and if, as Arrow and Nelson assumed, it only led to first-mover advantages, it would indeed be too risky for generous industrial support. Second-mover advantages are much more important, and more certain. What are they?

Yet, across the globe, there are hundreds if not thousands of other scientists, in universities, research institutes and company laboratories, publishing their findings on related problems in related fields, and if the company directed most of its scientists’ efforts at following those activities, the company would be much more likely to uncover a potentially useful advance. The time a scientist spends in the library or in the seminar room is more valuable to the company that employs him than the time he spends at the bench.

Second-mover advantages were worth nearly twice as much as first-mover advantages.

The biggest myth in science funding is that published science is freely available. It is not. Access to it is extremely expensive.

Companies are primarily interested in second-mover research; they need to know what everyone else is doing so that they can exploit advances from all over the world. The only people who can monitor other people’s research are the scientists. Companies, therefore, have to employ scientists. But scientists themselves are only really interested in firstmover research, and the best scientists are obsessed by their own, firstmover research. Yet even scientists need a salary, so companies and scientists agree a modus vivendi. Companies pay scientists to do the first-mover research that they,. the scientists, enjoy; while in return the scientists, through their reading and attendance at conferences, keep the company informed of developments worldwide. No irreconcilable conflicts of interest arise, because scientists cannot do good first-mover research without keeping abreast of developments, and the only people with the ability to keep abreast of developments in an area of science are the first-mover researchers in that area. But the financial consequences are that companies have to invest very heavily indeed in their researchers’ first-mover science to retain them as second-mover consultants.

Within a country, too, the linear model is also irrelevant, and scientists largely boost their national economic performances by means of the foreign research they capture. A series of recent reports have shown that much of the benefits of national R&D ‘spills over’ to foreigners. Moreover, the smaller the country, the more its economy depends on research performed elsewhere.

When national statistics for civil R&D are collected by the OECD, no immediate pattern emerges. But when the statistics are related to national GOP per capita, a strong correlation emerges: the higher the national GOP per capita, the higher the percentage of GOP devoted to civil R&D.

Those countries whose civil R&D is predominantly funded by industry spend more than those whose is predominantly funded by the state. This really does indicate that nationalisation reduces civil R&D budgets.

This longitudinal study thus confirms the following economic laws of civil R&D funding, as follows.

• The First Law of Funding for Civil R&D states that the percentage of national GDP spent increases with national GDP per capita.
• The Second Law of Funding for Civil R&D states the public and private funding displace each other.
• The Third Law of Funding for Civil R&D states that the public and private displacements are not equal: public funds displace more than they do themselves provide.

The discovery that the government funding of civil R&D disproportionately displaces private funding is desperately important. Governments, when they started to fund civil R&D, hoped that their funding would be additive (i.e. merely add to the amounts that industry already spent) but if that funding is displacive and, even worse, if governments actually displace more than they themselves provide, then the government funding of civil R&D damages the enterprise.

However, for civil R&D, the situation is self-correcting; it is being privatised by default. Prosperous countries are now devoting 2 per cent or more of GOP to civil R&D, and with tax limited to some 40 per cent of GOP, this forces a government intent on funding all civil R&D to allocating it 4 per cent of tax revenue. This is too much in the face of competing demands. Governments are not actually cutting budgets, but they are fixing them, and as the requirement for R&D rises with rising wealth, so effectively it is being privatised.

Academic science has, everywhere, largely been nationalised. Because it accounts for a much smaller share of GDP than civil R&D it will, moreover, tend to remain nationalised, unless governments lose their desire to keep it under control.

Science will attract generous patrons – if taxation spares them. Almost everyone in a wealthy country believes that pure science should be funded. Under dirigisme this sentiment is transmitted to the politicians (who, being human beings, will share the general values) but under laissez faire people, especially rich people, keep their money. The historical evidence shows that the empowerment of wealthy men by money breeds a sense of responsibility, which inspires them to endow science and universities. As a percentage of GOP (and corrected for the GOP per capita of the day) the empirical evidence shows that private donors are actually more generous to pure science than is the State.

Historically, the major beneficiaries of patronage have been, successively, the Church and the poor, the arts and sciences, and the secular charities. The historical record shows that the problem of private patronage for these beneficiaries has been overgenerosity, not underfunding.

We have constructed this book as a debate between Francis Bacon and Adam Smith. Who won? The world’s answer, of course, is Francis Bacon: The governments of all the industrialised countries now support their universities and their science much as Bacon prescribed, and if one asks about funding in a contemporary laboratory, university or ministry of science, the responses come straight out of the Advancement of Learning. To paraphrase, modem science policy is little more than a series of footnotes to Bacon.

Subsequent academic studies have vindicated Smith, not Bacon. The work of Mansfield, Gellman et al. and Langrish et al. has shown that at least 90 per cent of wealth-creating research emerges from applied, not pure science.

The work of Mansfield, Griliches and Rosenberg has shown that the free market drives companies into funding pure science, both in-house and within universities. Moreover, the historical evidence of the endowment of British and American universities and of their science by rich patrons under laissez faire disproves Bacon’s suggestion that governments need to do it.

If this book bas a message it is this: relax. Economic, technical and scientific growths are free lunches. Under laissez.faire they just emerge, like grass after the rain, through the efforts of individual entrepreneurs and philanthropists. Once the State bas initiated the rule of law and sensible commercial legislation, the goodies will flow – and laissez. faire is morally superior to dirigisme as it maximises the freedoms and responsibilities of the individual.

If you would like to learn more about role of scientific research in promoting progress, read my book From Poverty to Progress: How Humans Invented Progress, and How We Can Keep It Going.