Economics of Technological Change and Innovation Final
Carl Mackensen
Professor Vonortas
Economics of Technological Change and Innovation
Final Exam
Question 2
The maximum potential of an innovation requires its being taken up by the entire economy at large. In order to realize increased productivity everywhere, new technical processes must be focused on by producers. Positive externalities will result for the buyers and users of innovations. This, in turn, justifies the intervention in the market by policymakers. ““…In the history of diffusion of many innovations, one cannot help being struck by two characteristics of the diffusion process: its apparent overall slowness, on the one hand, and the wide variations in the rates of acceptance of different inventions, on the other.” (Rosenberg, 1972). It is quite expensive to examine the rate of take up of a new innovation at the overall aggregate level. Specific individual firms or consumers make the decision to take up the new product or service. It proves difficult to explain why some take this up sooner or later.
In terms of innovation adoption, market saturation takes place, “when there is complete displacement of older generation methods by a new technique or the displacement of earlier product varieties by the new types of goods and services. At this point the Schumpeterian notion of creative destruction is fully realized.” (Vonortas, 10/9/2023). In reality, incomplete saturation is more likely. To explain the decision making to take up new goods and services, we can examine the place of social norms, the way firms and customers are related, and how decisions come to take place.
E.M. Rogers, in “Diffusion of Innovations,” argued that there are five components which influence possible take up of the innovation. These include: “The relative advantage of the innovation, Its compatibility with the potential adopter’s current way of doing things and with social norms, The complexity of the innovation, and Trialability, the ease with which the innovation can be tested by a potential adopter, and Observability, the ease with which the innovation can be evaluated after trial.” (Vonortas, 10/9/2023). Rogers also argued that there are a number of exogenous aspects of society that may either aid or hinder adoption. These include, if the decision is a collective, individual, or central authority one, the channels of communication implemented to get information about the change such as word of mouth or media, the specifics of the social system within which possible take up is placed in, including such things as interconnectedness and norms, and how much marketing takes place.
While others focus on the external environment, Economists focus on the rational decision making of individuals. Specifically, they look at the aggregate of this, in which the person calculates the benefits of adoption versus the costs of change. Such a cost benefit analysis occurs in an environment of a lack of certainty with non-perfect information. The final decision takes place on the demand side, however it is the case that costs and benefits are usually influenced by supply side decision making. The final rate of diffusion takes place through adding up the decisions of the individual.
One way of construing diffusion of an innovation is the Epidemic Model. In this model, the spread of disease is replaced by the diffusion of technology. Here, two people have a random encounter within which one has already taken up the change, and the other hasn’t. Then, information changes hands detailing the innovation, which possibly can lead to the other person taking it up. “There is a fixed population of potential adopters N and all members of this population are identical in all characteristics. When a meeting occurs between an adopter and a non-adopter there is a fixed probability B that the current non-adopter will become adopter. The chance of such a meeting occurring depends on the proportion of the population that have already adopted the innovation D. Such meetings are random encounters, so the probability that an adopter meets a non-adopter is D(1-D). Then, the rate of adoption is dD/dt = BD(1-D)”. (Vonortas, 10/9/2023). Predictions can be garnered from this model. Specifically, the rate of adoption is bell shaped, like an S, adopter rate follows a logistic curve, and saturation takes place when all have adopted. This saturation depends on the likelihood that an encounter causes take up.
There are, however, limitations to this model. While the S shaped rate of adoption is proven, this rate has no basis in Economics. Further, people and firms aren’t always exactly the same. Another point is that not all decisions are rational. People may either take up the technology en masse, or resist. We cannot assume a static population of people to take up the tech. Further, while adoption is S shaped, we have no way of knowing the time scale.
There is another possible model we can use. Specifically, we can base it on the plausibility of the concept that tech diffusion depends on how much economic gain it endows people who take it up with, when considered in comparison with the status quo. “The advantage may consist of higher returns on investment through cost savings, increased sales, higher prices for an improved product, or a combination of these. It is an expected advantage representing the net present value of a future stream of monetary benefits, discounted by the probability that these benefits will actually occur. These probabilities are in reality the assessments of decision makers of the chances that a given decision will yield the anticipated results.” (Vonortas, 10/9/2023). As decision makers get more information about the tech change, we can assume an S shaped diffusion shape.
Stimulus-response models are yet another way of explaining the S shaped adoption curve. This allows for possible differences between firms, which in turn explain the way tech diffusion takes place. Stimulus-response questions focus on which aspects of firms and/or their environment explain when firms adopt, as well as information such as on the market or technical information must be internalized ahead of adoption. The stimulus is information, while firm attributes influence the degree to which they adopt.
There are a number of aspects that influence how an innovation diffuses through society. These include: “Characteristic of the Innovation, such as its origin, effect on inputs, place in the established production scheme, changes in the innovation, and complementariness among innovations. Characteristics of Potential Adopters, including tech specificity of the existing system, the firm’s financial position, tech capability, market position and alternative strategies, managerial attitudes, age of firms and industries. Characteristics of the Diffusion Process, such as external and internal information, external interests in diffusion, international diffusion. Characteristics of the Institutional Environment, including the patent system, laws and government regulations, specification writing agencies, insurance companies, and so on, and labor unions.” (Vonortas, 10/9/2023)
Factors on the supply-side which influence diffusion rates include the continuity of inventive activity, improvements in inventions after their first introduction, the development of technical skills among users, the development of skills in machine-making, complementariness between different techniques, and improvements to old technologies as well as the institutional context. (Vonortas, 10/9/2023)
Lastly, there are network effects and lock-in effects which influence tech diffusion. Network effects refer to the gains that one realizes based on the decisions of others to adopt. It is possible that not the best tech will be adopted, as network effects can lead to the adoption and maintenance of inferior tech based on it’s becoming a standard or universally adopted. Lock-in is based on their being a cost associated with changing tech.
Network externalities and standards are related, in so far as standards cause a proliferation of things which inculcate network externalities. “A technological standard increases the probability that communication between two products will be successful. Standards ease learning and encourage adoption when the same or similar standards are used in a range of products. A successful standard increases the size of the potential market for a good, which can be important in lowering the cost of its production and in increasing the variety and availability of complementary goods. “ (Vonortas, 10/9/2023).
In the end, products live in an ecosystem of their own making. The S shaped curve details the ways in which technology adapts, grows, replaces itself, and promotes economic growth. Such adoptions of technology take on characteristic and similar traits, so as to make the S shaped curve a reality. The diffusion rate can be construed as many things, such as an epidemic, or a natural word of mouth phenomenon. The core question becomes, when does an individual, firm, or country, jump from one S curve to the next? This is often dependent on a whole host of reasons. Most prominently, you need to have the installed capacity of the tech to be at a certain minimum level. Additionally, the new tech must prove to be a marked increase in betterment than the tech it replaces. To jump either too early or too late involves significant costs, and punishments. Therefore, it is vital to time things correctly. In order to maintain technological dominance, there should be a ready stream of potential new technologies available to the decision makers. Lock in effects can be significant hurdles to tech adoption, but there is also the potential for leap frogging. We must keep one eye to the future, when evaluating where we are.
Question 3
The Neoclassical Growth Model begins with a simple aggregate production function. Y = A f(K,L), where Y is GDP, K is capital stock, L is the size of the employed workforce. And A is the level of technology. The aggregate specific production function modifies this slightly to Y = A K^a * L^(1-a). Increasing K or L individually will increase Y, but this takes place at a diminishing rate. There are diminishing marginal products of K and L. The level of tech in the second equation is exogenous, and can be either increased or decreased. An increase in technology increases the value added for a particular input. Likewise, a innovation in the process of production causes an increase in the value added, for both firms and more generally.
We can rewrite the second equation as output per worker. This takes the form of y = Y/L = (A K ^a * L^(1-a))/L which in turn becomes (AK^a)/L^a = Ak^a where k = K/L. This can be interpreted as output per individual is based on the degree of tech and capital per worker. Another vital aspect is the accumulation equation. Accumulation can only benefit the stock of capital. S, or savings, is a constant proportion of output and is used for capital investment. A portion of this investment goes towards replacing obsolete and depleted capital, known as depreciation or d. Overall investment is equal to savings in a closed economy. For capital to grow, depreciation must be lower than savings. Labor, on the other hand, can not increase or accumulate. However, the model does assume a stable rate of growth which is exogenous. The core assumptions are given by the following equations. First, as we discussed, y=Ak^a, which means that output is based on technology level times capital to the power of a, as well as dK/dt = sY – dK which shows that the change in capital over the change in time is equal to savings time output minus depreciation times capital.
A further equation can be found, which is dK/dt = sy – (d+n)K. This can be interpreted as the change in capital per worker is equal to the savings per worker minus the depreciation and dilution component. The growth of capital per worker equals total investment per worker, minus the investment required for depreciation as well as to equip new workers. Given that we have the production function and accumulation equation, both in per worker terms, we can solve. In sum, the core outcome of the neoclassical model is that the economy comes to become a steady-state of capital and output per worker. An important corollary of this is the observation that the greater the economy is below the steady-state level, the quicker its growth rate of output per worker will be.
Increasing the savings rate results in a higher capital impact per worker and increased labor productivity. If we allow individuals to determine the level of savings, the final analysis of the model does not change. The economy comes to reach a steady-state of output per worker, or per capita. This is because of the diminishing marginal product of capital. When technology increases the output per capita grows, regardless of whether capital per capita remains the same. It will also increase the marginal product of capital.
Baumol, Blackman, and Wolff (1994) offer some additions which the preceding theory has neglected. Firstly, they offer the convergence hypothesis, which states that when one country is more productive than another, due to discrepancies in their ways of production, the worse off countries which are close to the leaders can begin a process of catching up. Some will even succeed. Further, ‘the catch up process will continue as long as the economies that are approaching the leader’s performance have a lot to learn from the leader.” As long as there is something to be learned by the worse off country, they will continue to grow until they have fully internalized all of the innovations which the leader has. There are also countries which are significantly behind the leaders. It is not practical for them to focus on the innovations of the leader, and these countries could fall victim to falling even further behind.
There are some criticisms of all of these. Today, there is significant inequality within advanced nations. There is inequity both within and between nations. Further, there is sample bias in the convergence theory. The countries examined for this are ex post successful. The chosen countries are chosen on the criteria of whether or not they have achieved economic prosperity towards the end of the period being examined, as opposed to their situation at the start. Looking at the industrialized, middle range, and low income countries, it has been shown that there is convergence between the first and second, but the latter falls further behind. The reasons given were ‘lack of education and impeding social arrangements that tend to swamp the advantages of backwardness.” (Baumol, Blackman, and Wolff, 1994).
There are other models available to explain growth discrepancies. Innovation is one such topic. Here, there is an upgrade in resources, as well as structural change. As different stages, there are different absorptive capacities. Further, in early stages, autocratic regimes do better than democratic ones. There is a need to consume less and invest more. How do we put productive activity to good use? There is imitative entrepreneurship, on which other nations are imitated, but it is new to the country in question that is developing. Further, just because a country gets close to the top players doesn’t mean that it is guaranteed to stay there.
Abramovitz (1994) notes why leaders are slower than followers in terms of growth. They have already invested in things. It takes time to change. The behind have less capital, so there is an opportunity for capital investment. The laggards move low productivity into high productivity, often into cities. Further there is the question of social capability. Growth is a ‘socio cultural economic technological phenomenon” (Vonortas, //2023). There is the need to entice multinationals to invest, whether with tax relief, cheap labor, well trained labor, or a large market. Lee and Malebra (2018) point out the middle income trap versus low income trap. Few manage to complete close the gap with the rich. There is a need to dramatically change policy and transform. Many fail. Further, catching up doesn’t mean copying. As they get closer, they need to change. For the neoclassical model, the mainstream view is that intervention by government is justified where markets fail. Public goods are one such example, and knowledge has aspects of being a public good. Further, serious positive externalities see under investment. Further, concentrated markets are not innovative, which also warrants intervention. People are assumed to know productive techniques and be rational. This is not always so, we see bounded rationality being exhibited. Those catching up need capability building and institution building through creative innovation systems. Lastly, when a nation enters the market is critical. Laggards enter late, and suffer a disadvantage, but they can offset this with cheap resources or labor.
Cicera and Maloney (2017) outline the ‘innovation paradox’ in the first chapter of their work for the World Bank. They argue that developing countries do far less innovation than do their developed counterparts. They cite Pritchett (1997) who found a ‘Great Divergence’ over the last two hundred years. The poor don’t catch up, while the rich continue to grow. This is argued to be due to the discrepancies between the rate of adoption of innovation between rich and poor. The capacity to ‘identify, absorb, and adapt technologies…is indeed a key part of the divergence story.” (Cicera and Maloney, 2017, page 3). Countries that may one hundred years ago have started with the same general economic conditions had significantly different capacities to innovate. Countries which have historically been unable to innovate and use technological development to their existing firms are also not likely to do so for newer industries and ventures. Therefore, innovation capacity seems to be the more vital point for economic development.
To me, I find these arguments to be interesting when compared with the neoclassical description of traditional growth analysis. I turn as an example to the state of the world at the end of World War II compared to today. Germany and Japan were decimated, with the USA and Russia being the only two viable nations left standing. While the USA continued to grow by virtue of pro growth and neoclassical development it also invested deeply in innovation. Russia maintained a system of centralized economic planning, which lead to its stagnation and eventual collapse in 1989. Germany and Japan, on the other hand, came back from the edge of non existence in terms of development and became, by the end of the century, two of the richest countries in the world. How did this happen? Through an emphasis on education, human capital development, science, and innovation. Further, they also pursued perfecting existing technology rather than costly experiments into the completely novel, leaving this for the USA and other rich countries. Using these methods, both have become economic powerhouses, and lessons that the developing world of today could learn from.
Question 4
Chapter seven of Gregory Tassey’s book “The Innovation Imperative” (2007) makes a number of claims about the way that technological development takes place. “As the engine of long-term economic growth, technology drives the creation of entirely new industries, adds substantial value to the economy, but eventually becomes obsolete and loses its value”. (Tassey, 2007, page 180). This is the way that technological diffusion occurs within a society. The “installed base effect” causes industries that dominate one cycle to rarely succeed in the following one. For the 40 years after WWII, the US dominated the technological economy sphere. During this time, dealing with life cycles was not a priority. However, this is no longer the case. As a result, life cycle analytics is an increasingly important component of dealing with economic growth. There are major cycles, and minor ones within those cycles. For long term growth to be sustained, the attributes that influence the S shaped growth curves must be understood.
Life cycles are accelerating, but there is constraint on innovation. Business concerns of old and new companies are different. “High tech firms are concerned with amount and type of government R&D funding, IP rights, cost of risk capital, and availability of scientists and engineers and…skilled workers. Older…industries…cite taxes, regulation, health and pension costs, trade barriers and tort laws as their most serious problems.” (Tassey, 2007, page 181).
To manage the attributes of long-term economic growth, we must recognize that product cycles are nested within larger technology life cycles which in turn construct a major technology cycle. A series of product cycles can come to fruition from an underlying general foundation of technology. For all subsequent technology cycles, aspects of the tech grow towards routinization and standardization, with consequential slowing of change, showing the approach of the exhaustion of the generic technology’s possibilities. As a life cycle reaches its natural end, competition shifts from significant changes in products to smaller, incremental ones as well as innovations in the means of production, or processes. Price-based competition becomes more the norm.
When we examine a life cycle, we can find that tech changes and innovation occur with fits and starts as time passes, mostly due to the occasional betterment of the overall tech. This does not mean, however, that when one cycle ends another simply begins. There is a demand side effect in which an older tech is still needed and necessary. As such, we have what is construed as a nested life cycle. Life cycles begin at differing origination points in the major overall lifecycle, before they are completely obsolete and replaced by new technologies.. When the generic technology reaches saturation and is available to all, tech change occurs at the level of products. Indeed, it is the case that the over all technology needed for each part of a larger system have to be open to use by all to facilitate multiple streams of innovation which in turn push forward the over all system.
By the time a technology comes to the middle of its saturation progress, and correspondingly large markets with a decidedly defined structure, larger firms with more R&D resources can facilitate significant general research. New tech, on the other hand, typically suffers from underinvestment, as firms see it as risky.
The overall tech life cycles are quite significant and relevant to the long term economic growth of an area, as they facilitate a progression of nested cycles which incorporates a whole field of similar tech development. It is also the case, however, that shifting from one cycle to another can be particularly onerous. “The length of a major cycle and the competitive position of the domestic industry over such cycles are particularly vital for general purpose technologies due to the fact that they create a whole ecosystem of innovative industries with huge economic impact. Unfortunately, global leverage by an initially innovative domestic industry is usually not continued over an entire technology life cycle.”
Lee and Malerba offer a different interpretation of the way technology changes. Changes in industry leadership is known as the ‘catch-up cycle’ take place with the passage of time, within a subset of the economy. During these cycles, laggards come forward as the would-be leaders and the established status quo leaders are usurped. This process repeats itself. Using a ‘sectoral system framework … identifies windows of opportunity that may emerge during the long run evolution of an industry.” (Lee and Malerba, page 338, 2016). Three such windows exist. The first deals with changes in knowledge and technology. The second, changes in demand, and the final, changes to public policy and institutions. A window opening, coupled with the response elicited among the latecomers and laggards impacts both catch up and leadership. Different sectors are of course different, with specific examples being studied, including, “mobile phones, cameras, semiconductors, steel, mid-sized jets, and wines.” (Lee and Malerba, page 349, 2016)
There are four stages to the catch-up cycle, including entry, gradual catch-up, forging ahead, and falling behind.” (Lee and Malerba, page 349, 2016). Leapfrogging can take place in which one actor adopts newer tech than its rivals. This is a component of the forging ahead state. There are many windows of opportunity that come to be frequently and surprisingly. They conclude on a ‘Schumpeterian’ view, as ‘exploiting a technological window is very critical to forging ahead.’ (Lee and Malerba, page 349, 2016). Further, windows can be competence-enhancing or competence destroying in terms of tech development. Further, we must also look at capabilities and strategies of both the advanced and laggards. New tech often coexists with existing tech, at least for a time. Alternatively, rapid change from one schema to another can also take place (as with cell phones). Laggards must capitalize on windows, and ‘build sector-specific capabilities that support actors, networks, and institutions.” (Lee and Malerba, page 350, 2016). However, this may be time intensive. The developed, on the other hand, must focus on lock-in or traps. They should attempt to maintain there advantage with others, and even improve it through innovation. A laggard could become stuck in the ‘middle income trap’ whereby they fail to ‘upgrade to high value-added products and is confined to performing activities with low value in the global value chain.” (Lee and Malerba, page 350, 2016). These countries should follow a set of policies that builds capabilities for innovation to take full advantage of the opening of such a window of opportunity, and also develop systems in which the full embrace of catch up innovation is possible.
The most profound thing a laggard country, industry, or firm can do to escape the middle-income trap is investment. I do not mean this strictly in the capital sense, though this is important, as is illustrated by the neoclassical growth theory. What I mean, as alluded to above in my second essay, is invest in fruitful innovation education. Through this, they have the potential to leap frog the leading countries and adopt tech that is substantially better, more developed, or more efficient. The use and deployment of both cell phones and solar energy are good examples. Countries such as Kenya and India have, at least since the end of World War II and the fall of colonialism, been laggards in their adoption of technology. Living a primarily agrarian existence, their growth was anything but assured. First came the boom of mobile telephony and the internet.
Rather than install land lines in huts, people in Kenya adopted cell phones. Not only this, but they were among the first to use cell phones as a means of mobile banking in low income countries. They developed a system by which small amounts of money could be paid from one person to another through their cell phones, thus revolutionizing the entire financial industry of the country, and allowing many small businesses to grow and flourish. This had further effects of increasing their international trade.
India, on the other hang, faced a similar scenario, but with electricity provision. Over the last 20 years the Climate Crisis has grown in significance to the entire world. It is now generally accepted that we cannot continue with business as usual, as concerns our emission of green house gases. When India was told that it could not develop the same way that all the Western industrialized countries had, it claimed hypocrisy on their would be international minders. Instead of moving forward completely with greenhouse gas emitting electrification, however, during the Paris Climate Accords, Al Gore made a fateful deal that would share the tech of Western PV solar companies with India.
In brief, countries can leap frog their former leaders by adopting new tech in novel and useful ways. This is based in the fundamental sharing of methods and innovations, as well as the absorptive capacity of the country in question. In order to full capitalize on the tech change window, a unit must be prepared. For a nation, this means having a populace that is well educated, particularly in terms of science and tech. As Benjamin Franklin said, “Luck is opportunity meeting preparedness.”
References
Vonortas, Nicholas. Economics of Technological Change and Innovation, 2023
Baumol, Blackman, and Wolff, 1994
Abramovitz, 1994
Cicera and Maloney, The Innovation Paradox, 2017
Tassey, Gregory, The Innovation Imperative, 2007
Lee and Malerba, Catch-up cycles and changes in industrial leadership: Windows of opportunity and responses of firms and countries in the evolution of sectoral systems, Research Policy 46 (2017)