SUMMARY
ECONOMICS OF INNOVATION
GEO-2211
,Contents
Lecture 1 innovation (micro-level)............................................................................................................................2
Lecture 2 diffusion (micro-level)...............................................................................................................................6
Lecture 3 knowledge (meso-level)..........................................................................................................................11
Lecture 4 industrial dynamics (meso-level)............................................................................................................14
Lecture 5 economic growth (macro-level)..............................................................................................................19
Lecture 6; Institutions (macro-level).......................................................................................................................23
Lecture 7; standardization and green public procurement....................................................................................28
Lecture 8; innovation in services.............................................................................................................................37
The past...............................................................................................................................................................37
The present.........................................................................................................................................................38
The future............................................................................................................................................................39
1
,LECTURE 1 INNOVATION (MICRO-LEVEL)
Industrial revolution let to growth
Importance technical change Mass production, larger scale, more visible
What’s innovation?
- New combinations of existing resources
- Invention is the first occurrence of an idea
Role innovation in eco and social change; Innovation is the engine of growth
Innovation introduces variety, after variety comes selection
Ways of analyse
There’s a neoclassical approach
- Technology is an exogenous shock (comes from the outside)
- Technology as information
and an evolutionary approach
- Technology is endogenous
- Technology as knowledge
Main evolutionary process
1. Generation of novelty: Perturbations which increase variety.
2. Selection: Mechanisms that reduce this variety systematically.
3. Preservation and transmission: Inertial forces which give continuity to the system.
Biological vs technological evolution
Difference biological vs technological evolution is the scope of change. The main similarity; interdependencies.
- Main difference: scope of change
o Biological variation occur through myopic random mutation ‘’by nature’’
Myopic is that you don’t know were they are going, trial & error
o Technology evolves myopically as a consequence of computational complexity
- Main similarity: interdependencies:
o The effect of mutations on a reproductive (technological) success is a function of both
selection conditions and structure of interdependencies
NK model
- Describes how firms use search strategies to innovate and how search strategies relate to the
complexity and the modularity of the innovation.
- Way to model interdependency
- Technological innovations are complex systems with interdependent components
- If you change one component, it affect others as well. You cannot act on component in isolation.
- Trade-offs; improvement in one component may damage others.
- Trial and error process
- Interdependency is key finding dimension of complexity
- There is a bounded rationality, we don’t know everything about the combination. We search local, a
bit every time
- Solution you will find is path dependence.
- You will end up with sub-optimal solution
- Global optimal is the best solution but is not always in your reach
o Gloabal highest w, local second highest
2
, Possibility space
- NK model is to understand how to navigate in complexity
- N is the amount of components in a technology
- An is the amount of possible states for component N
- The amount of possible designs S of a system is
- If every component has 2 possible states (An=2), the possibility space is binary and has size S=2^N
- E.g. a lock with 5 numbers/elements with 10 options; S= 10^5
Fitness landscape
- Each N has a ‘fitness’ score w
- A design’s fitness value W is the average fitness value of its components
- Components w may be correlated depending in the complexity.
- K is the amount of components that affect the fitness of each component (0<K<N-1)
- K is the way components are linkes
- When k=0, no components are affected by each other and there can be only one global optimum.
- K=2 when 2 components are affected. You can also have multiple local optima because you have
bounded rationality.
- Highest W global optimum
- K is parameter which tells the level of interdependence
- If you change one component and 2 are affected k=2
Question example
What is the likelihood (in percentages) that a firm searching in this landscape will end up designing a product
that corresponds with an optimum?
100%. You always end up at either a local or a global optimum. They are both optima.
What is the likelihood (in percentages) that a firm searching in this landscape will end up designing a product
that corresponds with the global optimum? What is its basin of attraction?
- The presence of a local optimum indicate that K>0
- The basin of attraction is the set of points from which it is possible to end up in the global optimum
one mutation at the time. In this case the global optimum is 111 (.87). 101, 011, 110 and 100 lie in its
basin of attraction.
3
ECONOMICS OF INNOVATION
GEO-2211
,Contents
Lecture 1 innovation (micro-level)............................................................................................................................2
Lecture 2 diffusion (micro-level)...............................................................................................................................6
Lecture 3 knowledge (meso-level)..........................................................................................................................11
Lecture 4 industrial dynamics (meso-level)............................................................................................................14
Lecture 5 economic growth (macro-level)..............................................................................................................19
Lecture 6; Institutions (macro-level).......................................................................................................................23
Lecture 7; standardization and green public procurement....................................................................................28
Lecture 8; innovation in services.............................................................................................................................37
The past...............................................................................................................................................................37
The present.........................................................................................................................................................38
The future............................................................................................................................................................39
1
,LECTURE 1 INNOVATION (MICRO-LEVEL)
Industrial revolution let to growth
Importance technical change Mass production, larger scale, more visible
What’s innovation?
- New combinations of existing resources
- Invention is the first occurrence of an idea
Role innovation in eco and social change; Innovation is the engine of growth
Innovation introduces variety, after variety comes selection
Ways of analyse
There’s a neoclassical approach
- Technology is an exogenous shock (comes from the outside)
- Technology as information
and an evolutionary approach
- Technology is endogenous
- Technology as knowledge
Main evolutionary process
1. Generation of novelty: Perturbations which increase variety.
2. Selection: Mechanisms that reduce this variety systematically.
3. Preservation and transmission: Inertial forces which give continuity to the system.
Biological vs technological evolution
Difference biological vs technological evolution is the scope of change. The main similarity; interdependencies.
- Main difference: scope of change
o Biological variation occur through myopic random mutation ‘’by nature’’
Myopic is that you don’t know were they are going, trial & error
o Technology evolves myopically as a consequence of computational complexity
- Main similarity: interdependencies:
o The effect of mutations on a reproductive (technological) success is a function of both
selection conditions and structure of interdependencies
NK model
- Describes how firms use search strategies to innovate and how search strategies relate to the
complexity and the modularity of the innovation.
- Way to model interdependency
- Technological innovations are complex systems with interdependent components
- If you change one component, it affect others as well. You cannot act on component in isolation.
- Trade-offs; improvement in one component may damage others.
- Trial and error process
- Interdependency is key finding dimension of complexity
- There is a bounded rationality, we don’t know everything about the combination. We search local, a
bit every time
- Solution you will find is path dependence.
- You will end up with sub-optimal solution
- Global optimal is the best solution but is not always in your reach
o Gloabal highest w, local second highest
2
, Possibility space
- NK model is to understand how to navigate in complexity
- N is the amount of components in a technology
- An is the amount of possible states for component N
- The amount of possible designs S of a system is
- If every component has 2 possible states (An=2), the possibility space is binary and has size S=2^N
- E.g. a lock with 5 numbers/elements with 10 options; S= 10^5
Fitness landscape
- Each N has a ‘fitness’ score w
- A design’s fitness value W is the average fitness value of its components
- Components w may be correlated depending in the complexity.
- K is the amount of components that affect the fitness of each component (0<K<N-1)
- K is the way components are linkes
- When k=0, no components are affected by each other and there can be only one global optimum.
- K=2 when 2 components are affected. You can also have multiple local optima because you have
bounded rationality.
- Highest W global optimum
- K is parameter which tells the level of interdependence
- If you change one component and 2 are affected k=2
Question example
What is the likelihood (in percentages) that a firm searching in this landscape will end up designing a product
that corresponds with an optimum?
100%. You always end up at either a local or a global optimum. They are both optima.
What is the likelihood (in percentages) that a firm searching in this landscape will end up designing a product
that corresponds with the global optimum? What is its basin of attraction?
- The presence of a local optimum indicate that K>0
- The basin of attraction is the set of points from which it is possible to end up in the global optimum
one mutation at the time. In this case the global optimum is 111 (.87). 101, 011, 110 and 100 lie in its
basin of attraction.
3
