Innovation. A review of Manso's paper "Motivating Innovation"

Seminar Paper, 2018

22 Pages, Grade: 1,0



1 Introduction

2 Methodology
2.1 Theoretical Approach
2.1.1 Exploration and Exploitation
2.1.2 The Principal-Agent Problem
2.1.3 Incentives for Exploration and Exploitation
2.1.4 Role of Commitment
2.1.5 Optimal termination policy of the principal for exploration
2.1.6 Provision of feedback
2.2 Reference to Empirical Implications and Approaches
2.2.1 Empirical Implications
2.2.2 Empirical Approaches

3 Critics and Further novel research

4 Conclusion

5 Appendix
5.1 Propositions
5.2 Figures

1 Introduction

In the age of disruptive technologies and business models, the most crucial factor innovation is moving more and more into the focus of attention of society. As leading innovative countries set the bar on efficient solutions higher, the determinants stay somewhat nebulous. While Zahra and Covin (1994) suggest that "Innovation is widely considered as the [life blood] of corporate survival and growth" Bessant et al. (2005) emphasize a more general definition: "Innovation represents the core renewal process in any organization." The latter statement suggests that innovation is not solely important for business organizations but rather for any kind of complex institutions. While the USA emphasized the significance of innovation by launching the Department of Innovation already in 2008, Germany has increased the public funding of Research and Development by 0.5 percentage points to 3.5 percent in the new coalition agreement in 2018. The Department of Trade of the UK exemplifies the relevance of accumulated innovation within a country: "If UK-based companies fail to innovate, jobs and profits will suffer, and our standard of living will fall compared with other countries." Blair (2003).

Many scientists have researched on the subject innovation. So do Aghion and Tirole (1994) argue that the outcomes of innovation are unclear, ergo it is difficult to find a cohesive contract which motivates innovation. Whereas the latter paper is studying the measurability of innovation, Manso’s article looks closely at the process of innovation and focuses on the central trade-off between exploration and exploitation. Moreover, Weitzman (1979) uses a simple bandit-problem to describe the innovation process.

The substantial difference to Motivating Innovation is that Manso does not only study individual decision problems but also complements the agent by the principal to embed the bandit-problem into a principal-agent framework. The Principal-Agent setting allows him to inquire the tension between exploration and exploitation under a more realistic contract in which the agent usually exhibits private costs when receiving tasks from the principal.

Recent papers give empirical support for some of the implications of the theories discussed. I will restrict the empirical findings to two of the tools which are essential practices to motivate exploration according to this paper, namely reward for early failure and continuation. For the reward of early failure, Tian and Wang (2011) find that firms which are financed by venture capitalists that tolerate failure are significantly more innovative. Regarding a continuation policy, Acharya et al. (2013) find that strict labor laws that restrict the dismissal of employees encourage firm-level innovation.

To complement the empirical findings, I suggest in Subsection 2.2 the empirical paper of Ederer and Manso (2013) which are directly concerned with the model proposed in this paper. They find that optimal exploration contracts, which are essential to motivate innovation, are indeed suited for tolerance and continuation. Exploration contracts described in Manso (2011) contain rewards in the form of a combination of stock options with long vesting periods, option repricing, golden parachutes and more. Whether these rewards empirically motivate innovation is not answered in this paper. However, Ederer and Manso (2013) find that golden parachutes foster innovation by giving the agent security in her insecure pursuit of exploration. However, in none of both papers, it is empirically stated, whether these contracts are more prevalent in innovative companies.

Furthermore, there is no suggestion on how to hire the right workers for the intended outcome. That is to say, how to design a contract that avoids conventional workers and shirkers but attracts creative workers if innovation is the intended outcome? The puzzle arises from the case in which the agent has superior knowledge about her type. The designed contracts could help solving the problem by serving as menus from which the agent shall choose according to her type. The self-allocation of menus reduces the information asymmetry by giving the principal more information about the agent’s type.

Lastly, Manso (2011) does not distinguish between the apparent difference in incentives of managers in private and public sectors. While earnings in public companies are transparent to the public, private ownership allows for more non-transparency. Implying the effect that public company managers could be more inclined to fulfill short-term expectations due to public pressure, while private company managers have the chance to agitate in a more covered setting. Ergo, private companies could be ceteris paribus more suited for innovation than public companies, as long-term commitment is crucial for an optimal exploration scheme.

This review of Manso’s paper Motivating Innovation is structured as follows. Section 2 sums up the theoretical and empirical results. While theoretical results are derived from this paper, empirical results make also use of other sources. Subsequently, in Section 3, I will firstly critically discuss the ambiguous effects of Job Security and afterward comment on some assumptions made in this paper while suggesting further novel research. Finally, in Section 4, I will conclude the review by stating the main findings of Manso (2011).

2 Methodology

2.1 Theoretical Approach

In this section, I summarize the theoretical approach of Gustavo Manso’s paper Motivating Innovation Manso (2011).

2.1.1 Exploration and Exploitation

In this subsection, I review the tension between exploration and exploitation. Manso defines exploitation as using conventional work methods to generate added value in a given task. On the other hand, exploration is the proxy for innovation, ergo new work methods in a given task. He approaches the problem with a two-armed bandit problem with one known arm, more specifically, probability of success. In this model, he uses a single-agent decision problem in two periods. This distinguishes from the Principal-Agent-Problem such that there are no external costs caused by the influence of a third party.

To differentiate between the two specifications, he enumerates each probability of successes for both actions exploitation and exploration in 1 and 2 respectively. In Action 1 the likelihood of success for both periods are equal, as conventional work methods do not change for both failure and success such that

Abbildung in dieser Leseprobe nicht enthalten

In Action 2 however, the conditional probability of success after success in the first period is higher than both expected probability of success in the second period and expected conditional probability of success after a failure in the first period as follows

Abbildung in dieser Leseprobe nicht enthalten

The different probabilities of the new work method derive from the assumption that the agent learns more about the outcomes after a given result of exploration. While Success is more of a positive signal, failure remains as a lesser positive signal for the second period.

If the agent succeeds while exploring in the first period, the expected conditional probability of success after succeeding in the first period is higher than both the probability of success in the first period while exploiting and the expected probability of success while exploring in the second period as follows

Abbildung in dieser Leseprobe nicht enthalten

As the agent is assumed to be risk-neutral, she has a discount factor of one. If she were risk-averse, her payoff would be less as the probabilities would decrease. In the agent’s maximization problem shown in regression 2, she chooses the action plan ijk which has the highest payoff. Where i 2 I is the first-period action, j 2 I is the second-period action in the case of success in the first period, and k 2 I is the second-period action in the case of a failure in the first period.

Abbildung in dieser Leseprobe nicht enthalten

The first line describes the expected payoff in the first period added to the following two lines of expected payoffs in the second period. Whereas the former line states the payoff after success in the first period and the latter one the payoff after a failure in the first period.

In a model with a team of agents or more periods, experimenting is more beneficial, and therefore the agents are willing to sacrifice more in the first period. Thus, it is possible that the agent is executing the second action in the second period, even though its expected probability of success is lower than the probability of success in both periods for the first action. With a team of agents, the discovery of a new successful work method by any agent will benefit the rest of the team. Additionally, a higher number of periods allows for more definite information from previous periods.

2.1.2 The Principal-Agent Problem

In this subsection, Manso expands the latter model of single-agent decision problem by the Principal-Agent Problem, in which the principal hires an agent to fulfill a given task. It is assumed that the principal does not observe the actions taken by the agent. So the principal offers the agent a menu of contracts w = {wF,wS, wSF, wSS, wFF, wFS} to incentivize the agent. Furthermore, the agent is now able to opt for shirking when given a task; thus there is a third defined action called Action 0.

Different from the latter model of the single-agent decision, private costs for the agent occur depending on the action chosen. Costs are for given actions i = 1, 2, 3 as follows:

Abbildung in dieser Leseprobe nicht enthalten

But the probability of success when shirking is lower than both other actions so that

Abbildung in dieser Leseprobe nicht enthalten

The motivation is to design the optimal contract for which the costs play an essential role. Such that as follows not only the expected payments from the principal to the agent are essential

Abbildung in dieser Leseprobe nicht enthalten

but also the incentive compatibility constraint which has to bind.

Abbildung in dieser Leseprobe nicht enthalten

The principal has to consider the IC in her profit maximization

Abbildung in dieser Leseprobe nicht enthalten

as the difference between reward and private costs of the agent ij must be the highest compared to the rest of possible constellations I'm.

To sum up, if c1 = c2 = 0, then there is no conflict between the principal and the agent, hence the principal solves the two-armed bandit problem in the last subsection. If on the other hand c2 = 1, it is so costly for the agent to employ the new work method that she will either exploit or shirk. So when c2 is very high, the principal-agent model studied here is not applicable to incentivize the agent to explore.

2.1.3 Incentives for Exploration and Exploitation

I attach the regressions and definitions of the following Propositions 1 and 2 to the Appendix because they are crucial to understanding the tension between exploration and exploitation, but leave Proposition 3 to 6 for the Online Appendix because their intuition is already given by both Proposition 1 and 2 and each following subsection.

In this subsection, I will review the incentives of the agent to exploit and explore. The first logical statement of Manso (2011) regarding exploitation is Proposition 1 in the Appendix. The following subsection is concisely summarizing Proposition 1.

Figure 1: The optimal contract that implements exploitation under the base-case parameters.

Abbildung in dieser Leseprobe nicht enthalten

This constellation is similar to pay-for-performance schemes, except for low costs of exploring in the case of wS, where the agent receives an extra premium for pursuing exploitation instead of exploration.

The intuition behind Proposition 1 is that the principal is trying to motivate the agent to exploit or in other words, to prevent the agent from exploring or shirking. Thus, the extra premium is decreasing in c2/c1, as increasing private costs of exploring diminish the incentive to pursue exploration as can be seen in the reward wS.

The second logical statement regarding exploration is stated by Proposition 2 in the Appendix. The following Figure 2 illustrates the optimal contracts in each situation of success and failure.

Figure 2: The optimal contract that implements exploration under the base-case parameters.

Abbildung in dieser Leseprobe nicht enthalten

This constellation is different from pay-for-performance schemes, such that the agent who succeeds early and fails later is rewarded lesser than when she fails early and succeeds then or even fails twice when c2/c1 is high.

The intuition behind Proposition 2 is that the principal must rather give incentives for later success than for short-term success and reward for early failure. For that, the principal must reward the agent higher in the case of a failure followed by success than a success followed by a failure, such that.


Excerpt out of 22 pages


Innovation. A review of Manso's paper "Motivating Innovation"
LMU Munich
Catalog Number
ISBN (eBook)
ISBN (Book)
innovation, disruption, motivation, economics, organizational economics, volkswirtschaftslehre
Quote paper
Oguzhan Bekar (Author), 2018, Innovation. A review of Manso's paper "Motivating Innovation", Munich, GRIN Verlag,


  • No comments yet.
Read the ebook
Title: Innovation. A review of Manso's paper "Motivating Innovation"

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free