This paper, which can be found here, adds to the literature on the relationship between competition and innovation, which has been the subject of longstanding attention by economists. However, existing empirical studies on competition and innovation suffer from a number of limitations.
The authors seek to address these limitations as regards a specific type of innovation models – so called ‘Step-by-Step Innovation Models’. Their study shows that, as long as key assumptions of the step-by-step innovation model are met, theoretical predictions of this model are confirmed by laboratory empirical data.
Section 2 looks at ‘Step-by-Step Innovation Models’.
The main characteristic of step-by-step innovation models when compared with previous Schumpeterian models (where competition is for the market) is that innovation incentives do not depend on post-innovation rents only, but rather on the difference between post-innovation and pre-innovation rents of incumbent firms. In the basic model setup, an industry consists of two firms which produce the same good and compete over selling the good to a customer. Firms can invest into technology, which lowers their cost of production. The crucial assumption at the core of all these models is that a “laggard firm” must catch up with a market leader before becoming a leader itself.
Thus, the model setup implies that, in an unlevelled sector, the leader gets the whole market. If one firm has better technology, it has lower costs of production and will lead the other company out of the market. If the firms are technologically at the same level, open price competition with no collusion will bring (neck-and-neck) firms’ profits down to zero. If the forms collude, they will earn a profit which will depend on the level of product market competition. Hence, an increase in the level of product market competition increases the innovation intensity of neck-and-neck firms. This is called escape-competition effect. Further, the intensity of innovation will depend not only on the level of product market competition, but also on the companies’ time-preferences. Patient neck-and-neck firms put more weight on future post-innovation rents after having become a leader, and, therefore, react more positively to an increase in competition than impatient neck-and neck firms.
The situation is typically different for firms which are technological laggards. If a laggard is very impatient, it will look at its short-term net profit flow if it catches up with the leader, which will decrease when competition increases (remember that perfectly competitive profits are zero). Since competition negatively affects the post-innovation rents of laggards, competition reduces rates of innovation by laggards (Schumpeterian effect). However, for low values of the discount rate, that is, for patient laggards, this effect is counteracted by an anticipated escape-competition effect: in other words, patient laggards take into account their potential future reincarnation as neck-and-neck firms, and therefore react less negatively to an increase in competition than impatient laggards (the composition effect).
Lastly, these theories predict that the higher the degree of competition, the smaller the fraction of neck-and-neck sectors in the economy. This is because more competition increases innovation incentives for neck-and-neck firms, whereas it reduces innovation incentives of laggard firms in unlevelled sectors. Consequently, this reduces the flow of sectors from unlevelled to levelled while increasing the flow of sectors from levelled to unlevelled.
Section 3 outlines an experimental study that tests the general validity of the predictions made by step-by-step innovation models.
This is done by reference to two different time-horizons: finite and infinite. This is because random assignment to levelled and unlevelled sectors is of utmost importance to cleanly identify the escape-competition and Schumpeterian effects. Further, to assess the impact of the rate of time-preference on behaviour, the experiment must feature infinite time horizons with varying stopping probabilities after each period. On the other hand, the composition effect relates to the long-run properties of a sector, so observing sectors for long and comparable periods of time is key here. However, this is very difficult to model with random stopping probabilities. Hence, a long and finite time horizon is best suited to assess the predictions.
Section 4 focuses on the infinite time-horizon experiment. This seeks to identify the escape-competition and the Schumpeterian effects.
At the beginning of the experiment, two subjects were randomly matched with each other, forming a sector. They subsequently interacted in a computerized step-by-step innovation game for an ex ante unknown number of periods. After each period, termination of the match was determined by the computer, based on a commonly known stopping probability. The timing of a period was as follows: First, one of the two subjects could make a costly R&D investment. Second, based on the investment, the computer randomly determined whether R&D was successful. If she was successful, the investing subject earned a point. Points were accumulated over all periods, and the point balance between subjects reflected the technological distance between the subjects. Third, payoffs were determined conditional on the point balance. Finally, the computer randomly determined whether the game would stop or another period would be played.
The experiment contained three different treatment variations. The first treatment variation was to randomly assign subjects to a levelled or an unlevelled point balance before the first period of the interaction. The second treatment variation changed the degree of competition when firms were neck-and-neck between no-competition and full competition. The third treatment concerned the time horizon: in some sessions, subjects faced a short time horizon—an 80% probability of ending the game after each period—while in other sessions the subjects faced a long time horizon—a 10% probability of ending the game after each period.
The results of this study were broadly consistent with the predictions of ‘Step-by-Step Innovation Models’, in particular: (i) an increase in competition leads to a significant increase in R&D investments by neck-and-neck firms and decreases the R&D investments of laggards; (iii) while an increase in competition leads to larger increases in investment in innovation in long time horizon scenarios as expected, the 4 percentage point difference which was observed is not statistically significant; (iv) short-term horizons lead to a decrease in investment by laggard firms by comparison to longer-term horizons.
Section 5 discusses the finite time horizon experiment, which analyses the composition effect.
The composition effect is about long-run predictions of these models, and hence requires observation of sectors over long and comparable periods of time. The setup of the experiment was similar to the one described in section 4 with some differences. First, the finite horizon experiment involved three different competition treatments: no competition, intermediate competition, and full competition. Second, all participants started level, but each subject was then randomly determined to invest in each period.
Again, the results were consistent with the prediction of ‘Step-by-Step Innovation models’. In particular, as competition increases, sectors become less likely to be neck-and-neck, and subjects are more likely to move technologically apart from each other.