Obstructive monitoring

• Date: 01 October 2020
• Published date: 01 October 2020
• Author: Aaron Finkle,Dongsoo Shin
• DOI: http://doi.org/10.1111/jems.12386

J Econ Manage Strat. 2020;29:873–891. wileyonlinelibrary.com/journal/jems © 2020 Wiley Periodicals LLC

|

873

Received: 10 February 2020

|

Revised: 25 May 2020

|

Accepted: 28 May 2020

DOI: 10.1111/jems.12386

ORIGINAL ARTICLE

Obstructive monitoring

Aaron Finkle

1

|Dongsoo Shin

2

1

Department of Economics, University of

Maryland, College Park, Maryland

2

Department of Economics, Leavey School

of Business, Santa Clara University, Santa

Clara, California

Correspondence

Dongsoo Shin, Department of Economics,

Leavey School of Business, Santa Clara

University, Santa Clara, CA 95053.

Email: dshin@scu.edu

Abstract

We consider a principal–agent relationship in which the principal's monitoring

can be obstructive to the agent, reducing the agent's productivity. We show

that, with obstructive monitoring, the optimal output schedule is distorted in

all directions—the high‐cost agent produces less, and the low‐cost agent pro-

duces more than the first‐best levels. Moreover, if the principal has a choice,

she will make monitoring deliberately obstructive, because when monitoring is

obstructive, although the agent's productivity decreases, his information rent is

extracted more effectively. We also show that obstruction is optimal even when

the principal is unable to commit to her monitoring strategy ex ante.

KEYWORDS

obstructive monitoring, principal–agent, rent extraction

JEL CLASSIFICATION

D82; D86; L21; L23

1|INTRODUCTION

Monitoring is a common instrument used by organizations to mitigate incentive problems. Supervisors oversee worker

activities to prevent shirking, firms are subject to regulatory oversight to assure laws are being followed, lenders look

over borrowers to prevent misappropriation of funds. Monitoring has a rich literature as to its benefits in various

principal–agent relationships. It is well understood by now that monitoring allows organizations to reduce information

rents and inefficiencies in the presence of asymmetric information. While monitoring has its merits, in practice,

monitoring can be intrusive and obstructive. Regulators often impose costs on those firms being monitored, such as

requiring them to provide documentation or slow down production to verify compliance with regulations. Supervisors

sometimes “bully”their subordinates while monitoring work effort. Many organizations monitor broad amounts of

employee activity, including e‐mails and internet use which can have negative impacts on the employee morale. In

general, monitoring may not only be costly for the party conducting it (the principal), but also may impose costs on the

party being monitored (the agent).

In extreme cases of obstructive monitoring, every single step a worker takes is dictated by the supervisor. It is

reported that in some organizations, a worker's internet access, online sessions, and electronic conversations are

monitored, as well as other actions performed by the worker.

1

These forms of monitoring can impose undue stress,

thereby lowering a worker's productivity and efficiency in performing tasks. Indeed, among practitioners, micro-

management now commonly refers to an organization's excessive control with the effect of obstructing processes.

Management styles, such as interfering oversight of a worker's effort, requiring burdensome workload reports, or

sampling a product line to analyze quality, can all obstruct and reduce workers' productivity. In practice, workers

frequently complain that intrusive monitoring actually lowers productivity, because it is hard to get things done under

such disruption.

2

Simply put, while monitoring enables an organization to keep its workers honest, it lowers their

productivity when obstructive.

While there is a long literature on principal–agent relationships focusing on the trade‐off between the principal's

cost of monitoring and the benefit of inducing the agent's truthful behavior, no considerations have been made of this

additional cost associated with monitoring incurred by the agent—the obstructiveness of being monitored. Our study

incorporates the obtrusiveness of monitoring by considering the monitoring cost borne by an agent. We adopt a hidden

information model in which the principal can monitor the agent, but at the same time, the agent's productivity is

lowered when he is monitored.

As is well known in the agency literature, in the absence of monitoring, the optimal contract exhibits “no distortion

at the top,”that is, the output level for the low‐cost agent is not distorted, while the output level for the high‐cost agent

is “distorted downwards”as a result of rent extraction. The conventional result in the literature is that introducing

monitoring always reduces the distortion in the optimal output schedule since monitoring extracts the agent's rent. Our

result suggests that, when obstructive, monitoring can increase distortions while extracting the agent's information

rent. While monitoring mitigates the agent's incentive to shirk by misrepresenting his type, we show that obstructive

monitoring generates output distortions in all directions—in the optimal contract, the output level for the high‐cost

agent remains to be below the first‐best level, but the output level for the low‐cost agent becomes above the first‐best

level.

In our model, the principal observes the output level, but without monitoring, she cannot observe the agent's

productivity (type). The agent reports his type to the principal which dictates whether the agent is subject to mon-

itoring. Although monitoring the agent's effort is costly for the principal, it enables the principal to learn the agent's true

type and dictate the agent produce the contracted level of output linked to his type. The extra cost of monitoring is that

it is obstructive—monitoring lowers the agent's productivity, thereby making it more costly to complete the task than

when not monitored. In other words, if obstructive monitoring takes place, the agent becomes less productive.

As mentioned above, when monitoring is obstructive, while the high‐cost agent produces less than the first‐best

level, the low‐cost agent produces excessively in the optimal contract. Since the low‐cost agent has an incentive to

misrepresent his type to shirk (thus reaping a rent), the principal conducts monitoring with a strictly positive prob-

ability only when the agent announces that he is the high‐cost type. If the low‐cost agent misrepresents his type as high‐

cost and monitoring takes place, then the agent must produce the output level for his true type. However, because

monitoring lowers his productivity, the agent bears a higher cost of producing the required output level for his

true type.

The chance of being monitored raises the expected cost to the low‐cost agent from misrepresenting type, much like a

penalty. By increasing the low‐cost agent's output level, the principal increases the agent's cost of misrepresenting his

type. And with a higher output level, the penalty for shirking by misrepresenting his type is larger (if monitored with

obstruction, the agent has to produce more at a higher cost), and thus costly monitoring is needed with a smaller

probability to induce the agent's truthful behavior. When monitoring is obstructive, distortions in all directions are used

as incentive devices to extract the agent's rent. This potential benefit, however, comes at a cost of lower productivity.

We then examine the principal's choice of whether to obstruct when monitoring the agent's productivity. The

principal can select from different monitoring technologies of varying levels of obstruction to the agent. We demon-

strate that, when monitoring is optimal, the principal will choose a technology which is strictly obstructive. In other

words, when monitoring the agent, obstruction is beneficial to the principal compared with a nonobstructive mon-

itoring technology. Although obstructiveness decreases the agent's productivity, it serves as an additional device of rent

extraction—the agent's information rent is zero regardless of his type when monitoring is obstructive. Therefore, our

result sheds light on intrusive micromanagement often practiced in organizations. While constant and detailed feed-

back that is often excessively focused on procedural minutia affects a worker's productivity negatively, it has its strategic

value.

In our base model, no explicit penalty is introduced to focus our analysis on the benefit and cost of obstructive

monitoring. We then relax this assumption to demonstrate that the obstruction strategy is robust, if the amount of

explicit penalty the principal can impose cannot be too large. The reason is that, the principal uses obstruction to

penalize the misrepresenting agent by transferring a proportion of monitoring cost to him—obstruction allows the

principal to reduce the occurrence of monitoring thereby reducing her expected cost of monitoring. Therefore, even

when an explicit penalty is available, for the amount of penalty small enough, obstructive monitoring remains optimal.

Lastly, we extend our analysis to the case in which the principal cannot commit to her monitoring strategy—

decisions for monitoring and obstructiveness are made after the agent reports his type. The principal may have no

874

|

FINKLE AND SHIN