Gregory Tassey (2001)

“R&D Policy Models and Data Needs”, 37-71, in: Maryann P.

Feldman / Albert N. Link (eds.): “Innovation Policy in the

Knowledge-Based Economy”. Boston / Dordrecht / London: Kluwer Academic Publishers.

Needed: A Technology-Based Economic Policy Model S&T / “S” can be more easily defined: “… a pure public good” (p. 37) “T” is more complicated: a “mixed” private

and public good: “The ‘T’ portion is another matter. Unlike basic science,

technology is a ‘mixed’ good, containing both private and public elements” (p. 37)

One problem with regard to “T”: often poorly defined policy recommendations:

“Most S&T analyses gloss over the economics of Research and Development (R&D) investment and the associated market failure mechanisms, jumping to a set of poorly defined policy recommendations” (p. 37)

Life Cycle Models of Technology Development (1) Technology life cycles and the transition between

life cycles: “To deal effectively with this complexity, decision makers need a

realistic and policy-relevant economic framework of investment and performance over the technology life cycle and, equally important, the transition between life cycles” (p. 39)

Drucker put forward the hypothesis of a 50 years “major technology life cycle” from basic research to market application (market introduction):

“Drucker argues that major technology life cycles have persistently taken about 50 years from the initiation of significant basic research to the emergence of market applications” (p. 39)

Life Cycle Models of Technology Development (2) Three different types of life cycles: moving

from the specific to the more general “The technology assessment literature identifies three distinct

cycles” (p. 39)

Life Cycle Models of Technology Development (3) (1) Product Life Cycle: “The shortest and least controversial is the ‘product life cycle’,

which is simply the time from concept or initiation of product development through market penetration and eventual decline” (p. 39)

(2) Generic Technology Life Cycle: “A number of product life cycles are typically derived from the

same underlying generic of fundamental technology, which collectively forms a ‘generic technology life cycle.’ The generic technology is not static but evolves during its life cycle” (pp. 39-40)

(3) Major Technology Life Cycle: “Eventually, a major new science base appears, allowing a

transition to a new long-term ‘major technology life cycle’ or ‘wave’” (p. 40)

Life Cycle Models of Technology Development (4) The transition from one generic/major life

cycle to another always defines a sensitive period/process of risk:

“The transition between two generic technology life cycles presents a different set of competitive threats” (p. 40)

Life Cycle Models of Technology Development (5) At the same time: also the transition from

basic research to technology (applied technology) research marks a period of risk

“However, a major problem for R&D policy arises at this transition from basic research to technology research. Here, for the first time, market risk assessments must be added to estimates of technical risk. …technology risk, with its ultimate objective of market applications, encounters an initial major increase in total risk because the scientific principles presented must now be proven capable of conversion into specific technological forms with specific performance attributes that meet specific market needs” (p. 41)

See here Figure 2: Risk Reduction over an R&D Life Cycle

15 years 10 years 5 years commercialization

R

R'

R''

Risk

Risk Reduction Over a Technology Life Cycle

Source: Gregory Tassey, The Economics of R&D Policy, Quorum Books, 1997, Chap. 7

Basic ResearchGeneric Technology

Research Applied R&D

R'''

Infratechnology Research

•

•

A

BOld Technology

New Technology

Potential or Actual Performance/Price

Time

1

2

Transition Between Two Technology Life Cycles

Source: Gregory Tassey, The Economics of R&D Policy, Quorum Books, 1997, Chap. 7

•A

BOld Technology

New Technology

Potential or ActualPerformance/Price

Time

Life Cycle Evolution: Generic Technology

B'

••

•

•

A

BOld Technology

New Technology

Potential or ActualPerformance/Price

Time

••

C'

C

Life Cycle Evolution: Infratechnology

Elements of an Economic Policy Model of R&D (1): Generic Technology (a) Definition of “generic technology”: general

technology, as a spin-off of university’s basic research, demonstrates its market potential

“Typically, before large amounts of private sector funds can be committed to developing market applications of the technology, the generic technical concept must be demonstrated. This generic technology research provides evidence that the general technology may work in specific market applications” (p. 47)

Generic technology as a prerequisite for massive private funding:

“The existence of a demonstrated generic technology as a necessary condition for huge amounts of follow-on private sector investment was originally identified by Nelson (1987, 1992), but policy models have only vaguely recognized its critical importance” (p. 48)

Elements of an Economic Policy Model of R&D (2): Generic Technology (b)

Generic technology (the maturity of generic technology) indicates later market potentials:

“The broader and deeper the generic technology is, the greater the amount of R&D stimulated and the larger the number of market applications eventually produced” (p. 48)

Biotechnology as an example for generic technologies:

“Biotechnology provides an excellent example of how generic technologies arise out of the science base, but ahead of specific market applications” (p. 48)

Elements of an Economic Policy Model of R&D (3): Infratechnology (a)

Definition of infratechnology: “The other category of industrial technology with significant

public good content[1] is infratechnology. Infratechnologies are a varied set of technical tools that perform a wide range of measurement, integration, and other infrastructure functions” (p. 50)

[1]) “Generic technology” also expresses a public good content.

Elements of an Economic Policy Model of R&D (4): Infratechnology (b)

Functions of infratechnology include: (1) Measurement and test methods: “Measurement and test methods” (p. 50)

(2) Artifacts: “Artifacts such as standard reference materials that allow

these methods to be used efficiently” (p. 50)

(3) Scientific and engineering databases: “Scientific and engineering databases” (p. 50) (4) Process models: “Process models” (p. 50)

Elements of an Economic Policy Model of R&D (5): Infratechnology (c) (5) Technical bases for interfaces between

components of systems technologies: “Technical bases for both physical and functional interfaces

between components of systems technologies, such as factory automation and communications systems” (p. 50)

“Research-intensive” infratechnologies are of a crucial importance for “technology-based” economic activities:

“… the complexity of technology-based economic activity and the demands by users of technology for accuracy and high levels of quality have reached levels that a large that a large number of diverse research-intensive infratechnology are required – even within single industries” (p. 50)

Market Failure and Underinvestment (1) A four-fold typology of R&D underinvestment: “In general, four basic categories of under investment can and

do occur” (p. 51)

(1) Aggregate R&D underinvestment: “Aggregate under investment by an industry (e.g., insufficient

total R&D)” (p. 51)

(2) Underinvestment in applied business R&D: “Under investment in applied R&D in new firms (e.g.,

insufficient venture capital” (p. 51)

Market Failure and Underinvestment (2) (3) Underinvestment in generic or major

technology life cycles / “Next generation technology”:

“Under investment in new generations of existing technology or in radically new technology (e.g., insufficient generic technology research)” (p. 52)

(4) Underinvestment in technology infrastructures / infratechnology:

“Under investment in supporting technology infrastructures (e.g., insufficient infratechnology R&D)” (p. 52)

Market Failure and Underinvestment (3) Technology life cycles: depending on the

status of the specific cycles specific market failures can occur and specific policy responses and cycle transitions are appropriate

“The concept of technology life cycles is particularly important for R&D policy because it implies a time order or evolutionary character … for various market failures that appear at different points in the typical life cycle. Hence, this concept helps determine appropriate policy responses within life cycles as well as the critical transitions between cycles” (p. 51)

Market Failure and Underinvestment (4) The cyclical character of technology development

often has the following implications: underinvestment and connected market failures have a tendency to repeat itself (with each new cycle)

“Because the process of technology development evolves cyclically, under-investment due to market tends to repeat” (p. 52)

Specific market failures demand specific industry and/or government responses:

“Moreover, distinctly different types of market failure exist and therefore require different government or government/industry response modes” (p. 52)

Market Failure and Underinvestment (5): Important Sources of Market Failure (a)

Major types of market failure in connection to technology life cycles (pp. 52-54):

“The following are the major types of market failure that can and frequently do occur over the typical technology life cycle” (p. 52)

(1) “Technology is inherently complex”; (2) “Increased global competition shortens

R&D and product life cycles”;

Market Failure and Underinvestment (6): Important Sources of Market Failure (b)

(3) “The benefits from technology tend to diffuse or leak to firms beyond the originator (innovator) – price spillovers and knowledge spillovers”:

“Price spillovers occur when the market price does not fully capture the additional benefits from the new technology”; “Knowledge spillovers refer to leakage of the innovator’s new technical knowledge horizontally to competing suppliers. This type of spillover is good for the economy as a whole, but it decreases the expected returns for potential innovators” (pp. 52-53)

Market Failure and Underinvestment (7): Important Sources of Market Failure (c)

(4) “Market structure can reduce expected rewards from technology investment”:

“An increasing number of technologies are systems. System components must interface seamlessly with other components to work effectively. …Interfaces between complex components frequently are just as complex as the components themselves” (p. 53)

(5) “Corporate strategies often are narrower in scope than a new technology’s market potential”:

“Some new technologies (advanced ceramics, for example) have applications in a number of markets previously served by very different technologies and hence industries. Companies in the existing industries typically do not have the strategic profile or the production and marketing knowledge to target all the potential applications (i.e., to capture economies of scope)” (p. 53)

Market Failure and Underinvestment (8): Important Sources of Market Failure (d)

(6) “Increasingly dominant ‘systems’ technologies require complex infrastructure”;

(7) “Market segmentation is an increasing problem for production technologies”:

“Sophisticated customers demand customized versions of the technology which require highly complex and flexible productions systems” (p. 54)

Market Failure and Underinvestment (9): Transitions between Technology Life Cycles (a)

Within a life cycle there is often “conventional market dynamics”: for efficiency improvements of a product life cycle the funding of the private sector is mostly sufficient

“Within each life cycle, technology evolves according to conventional market dynamics. Efficiency can be achieved within short-term product life cycles largely by the private sector, with modest infratechnology support from government” (p. 55)

Market Failure and Underinvestment (10): Transitions between Technology Life Cycles (b)

Transitions from one technology life cycle to another can imply major shifts in competitiveness leadership: even from one national economy to another

“One of the main factors causing shifts in market share leadership across companies, industries, and countries is the phenomenon of transitions between successive generic or major technology life cycles that serve the same market function (e.g., computing or communication)” (p. 54)

“However, significant market share shifts can occur across national economies, if the original generic technology is not improved over its life cycle. Even greater shifts in competitive position can occur between generic and major life cycles” (p. 55)

The Tenuous Nature of Competitive Advantage (1): Examples of Technology-Based Markets Lost to Foreign Competition (a)

There are several examples for primary U.S. inventions that turned into economic revenue success stories for Japanese companies (pp. 56-57):

“The following are just a few examples of technology-based economic growth opportunities that have been lost by U.S. industry for a variety of reasons” (p. 56)

Semiconductor production equipment (steppers); Flat panel displays; Advanced ceramics; Robotics;

The Tenuous Nature of Competitive Advantage (2): Examples of Technology-Based Markets Lost to Foreign Competition (b)

Videocassette recorder; Semiconductor memory devices; Digital watches; and Interactive electronic games.

The Tenuous Nature of Competitive Advantage (3): Under Investment in Generic Technology Research (a)

The testing of the general marketability of generic technology does not consume much private investment: however, it is crucial for the technology cycle and may lead to heavier follow-up investment for applied research and experimental development

“The next step – proving the generic technological concept, so that corporate R&D managers can make the technical and market risk assessments necessary for follow-on applied R&D investments – is much more complex. This early-phase generic technology research does not absorb a large portion of total R&D spending in most areas of technology… However, its critical position in the R&D cycle means it has the potential to leverage much greater amounts of follow-on applied R&D” (p. 58)

The Tenuous Nature of Competitive Advantage (4): Under Investment in Generic Technology Research (b)

University-industry-government partnerships with regard to the early phase of the testing of generic technology should compensate private risk and possible market failures:

“As the next section describes, various combinations of industry, government, and universities fund or conduct generic technology research through a variety of organizational arrangements. These partnerships are attempts to deal with the quasi-public good nature of generic technologies and the consequent set of market failures that result from the mismatches between their characteristics and private-sector risk tolerances, R&D capabilities, and market strategies” (p. 58)

The Tenuous Nature of Competitive Advantage (5): Under Investment in Generic Technology Research (c)

The dilemma for companies’ R&D strategies: incremental R&D generates the revenues and radical R&D the profits

“The critical point for R&D policy is that major technological advances take more time and entail more risk than do incremental change and improvements. …Yet, as a study by Kim and Mauborgne[1] … clearly indicates, longer-term and more radical R&D projects are more profitable” (p. 59)

[1]) W.C. Kim and R. Mauborgne (1997). “Value Innovation: The Strategic Logic of High Growth”. Harvard Business Review 75: 1, January-February, pp. 102-112

See here Figure 4: Profit Differentials from Major and Incremental Innovations

Next Generation Innovations

(14% of Launches)

Incremental Innovations

(86% of Launches)

62% of Revenue 38% of Revenue

61% of Profits

39% of Profits

Profit Differentials for Major and Minor Innovations

Source: W. Chan Kim and Renee Mauborgne, “Value Innovation: The Strategic Logic of High Growth”, Harvard Business Review, 1997

The Tenuous Nature of Competitive Advantage (6): Under Investment in Infratechnology Research

Firms are often cautious to invest too much into infratechnology: great benefit for the economy as a whole, but often only of limited value for the innovator

“Because of the large number of infratechnologies required by a single technologically advanced industry, the collective economic impact is large. However, their infrastructural and hard-to-visualize character along with the diffuse nature of their impacts creates difficulties for policy makers” (p. 60)

Policy Implications (1): The Relative Size and Impacts of Government-Funded Research

Demonstrating the marketability of a generic technology and the availability will lead to follow-up private investments in applied research and experimental development:

“… significant risk reduction in terms of advancing the generic technology to at least proof of concept or laboratory prototype often is essential to stimulate the much larger applied R&D investment by individual companies that eventually brings products, processes, and services to the market. The availability of appropriate infratechnologies makes this process more efficient and, in some cases, is necessary for it to occur at all” (p.61)

See here Figure 5: Relative Expenditures by Phase of R&D over Technology Life

Policy Implications (2): Matching Market Failure and Policy Response (a)

Possible policy responses for specific types of underinvestment (pp. 62-63):

“Four major categories of under investment were identified in Section 6.0. Each requires a very different policy response” (p. 62)

(1) “Aggregate under investment by an industry – insufficient total R&D” – policy response: tax incentives:

“Tax incentives can be effective as long as the R&D being targeted is comparable to the type already pursued by industry… In other words, a tax incentive can stimulate more of the same type of R&D already conducted” (p. 62)

Policy Implications (3): Matching Market Failure and Policy Response (b)

(2) “Under investment in the formation of new firms – insufficient venture capital” – policy response: a government role and policy is possible if the new generic technology is radically new, promises great social and economic benefits, and can be pursued also by smaller companies (thus is not that capital intensive) / for example biotechnology:

“Venture capital is plentiful in the United States and available for most areas of technology, once the generic technology is sufficiently advanced… The policy issue is how to get to that point. A government role in advancing generic technology research has been accepted in the United States only in a few situations… That is, if a new technology has potentially large social as well as economic benefits, is not capital intensive (it can be supplied by small firms), and is radical enough … then R&D policy can consider subsidizing the creation of a new industry structure as a policy objective. Biotechnology is an example of a new technology that meets these criteria” (p. 62)

Policy Implications (4): Matching Market Failure and Policy Response (c)

(3) “Under investment in new generations of existing technology or in radically new technology – insufficient generic technology research” – policy response: partnership programs and joint industry-government programs:

“Tax expenditures tend to leak to both technologies and phases in the R&D process that do not require government support. …Early-phase, generic technology research often is conducted through a variety of partnership mechanisms… All nations with technology-based growth strategies have industry-government programs to cooperatively advance the early phases of a variety of technologies with considerable economic potential” (pp. 62-63)

Policy Implications (5): Matching Market Failure and Policy Response (d)

(4) “Under investment in supporting technology infrastructures – insufficient infratechnology R&D” – policy response: government (public) laboratories

“[This] …argues for a strong role by government laboratories in the conduct and diffusion of infratechnology research” (p. 63)

Policy Implications (6): Responding to Technical and Market Risk Barriers (a)

Industry and government responses to market failures must be flexible, responding specifically to the type of market failure

“Therefore, both industry and government responses to market failures and resulting under investment in R&D and/or slower market penetration by the new technology also must vary” (p. 63)

Policy Implications (7): Responding to Technical and Market Risk Barriers (b)

Reluctance of firms to invest in early generic technology research: some factors / reasons

“Nevertheless, the decision process for investing in the early phases of technology research is a difficult one for individual companies to manage effectively. There are several reasons for that difficulty” (p. 64)

(1) Competition pressures consume the firms’ energies:

“Competitive pressures of fighting for market share in the current technology life cycle consume sorporate energies”(p. 64)

Policy Implications (8): Responding to Technical and Market Risk Barriers (c)

(2) Long-term R&D often demands different strategies than conventional R&D:

“Corporate management applies different decision criteria and metrics to long-term projects compared with the bulk of corporate R&D” (p. 65)

(3) Technological and market risks of new technologies:

“Recognition of the extremely high technical and market risks associated with a potential technology whose concept has not been proven even in the laboratory” (p. 65)

Policy Implications (9): Responding to Technical and Market Risk Barriers (d)

(4) Sometimes the competences of a firm are not broad or diversified enough for focusing on new technologies:

“R&D capabilities needed to launch a significant research program in a new technology, especially involving multiple research disciplines and laboratory facilities, are frequently incomplete with individual firms” (p. 65)

Policy Implications (10): Responding to Technical and Market Risk Barriers (e)

Risk-pooling, R&D-complementing, and cost-sharing research consortia play a crucial role for coping with the risk increase at the beginning of new generic technology research:

“In response, risk pooling is a common strategy for conducting early-phase technology research. Consortia are widely used for this purpose in advanced economies. A number of cooperative organizational forms can be used to share costs and complementary R&D capabilities… Cost-shared research consortia with various combinations of industry, government, and universities as partners are the efficient approach to conducting much early-phase generic technology research in the United States” (p. 65)

Policy Implications (11): Responding to Technical and Market Risk Barriers (f)

After the early-phase generic technology research the individual firms often are capable of continuing with conventional R&D strategies:

“Once beyond this early-phase, industry can usually apply conventional R&D decision criteria and take on most of the applied research and development required to achieve commercialization of the generic technology” (pp. 65-66)

Policy Implications (12): Responding to Technical and Market Risk Barriers (g)

For the purpose of applied research some firms also will form industry consortia: the duration of industry consortia typically is shorter than that of scientific research programs

“… firms also will use consortia for applied research. However, such industry-led consortia seldom have a life or more than five years, with three to five years being the typical range… Scientific research programs are usually at least 10 years long and often span several decades” (p. 66)

Policy Implications (13): Responding to Technical and Market Risk Barriers (h)

The great dilemma and challenge for R&D: the so-called “funding gap” between publicly funded basic research and privately funded applied research and development

“The widely discussed Ehlers’ report provides a Congressional view[1] of the ‘widening gap between federally-funded basic research and industry-funded applied research and development’” (p. 66).

[1]) U.S. House of Representatives, Committee on Science (1998). “Unlocking Our Future: Towards a New National Science Policy”. Report to Congress, Washington, DC.