Download StudyGeneral

Study Overview

An Evaluation of the Relative Efficacy and Effectiveness of Written and Verbal Contracts: Lessons from a Field Experiment in Urban Malawi
Study is 3ie funded:
Study ID:
Initial Registration Date:
Last Update Date:
Study Status:
In Development

Designing and enforcing the optimal contract is complicated and tedious, particularly in settings where formal contracts are not the norm. In many low- and middle-income countries, for example, transactions between customers (buyers) and firms (sellers) are often conducted informally to the extent that written receipts or other types of documentation are often not provided when purchasing goods in a market, negotiating prices for services, or disputing the receipt or quality of previous purchases. The introduction of a transparent agreement (either written or verbal) in informal settings could serve to mitigate unforeseen contingencies, limit the uncertainty on transaction outcomes, settle disputes over transactions between buyers and sellers, improve performance standards and accountability, and raise productivity. At the same time, however, a formal contract may limit the scope for renegotiation over the price of the good in future periods, particularly as reputation between buyers and sellers is built over repeated interactions. In this study, we propose to conduct a field experiment in which we test the relative effects of introducing and enforcing task-oriented contracting structures on outcomes related to contract compliance, levels of effort exerted by contracted agents, and service quality. The study will be conducted among a sample of 100 print shops, kiosks, and firms in Lilongwe, Malawi. We particularly are interested in testing the extent to which introducing certain imposing conditions (e.g. tight completion deadlines, strict printing specifications, etc.) along with a formal contract impact firm performance.

Registration Citation:
Economic Policy
Information and Communications Technology
Private Sector Development
Public Sector Management
Social Protection
Additional Keywords:
firms; contracts; printing; Malawi
Secondary ID Number(s):

Principal Investigator(s)

Name of First PI:
Mahesh Karra
Boston University
Name of Second PI:
Thomas Gautier
Kiel Institute, Boston University

Study Sponsor

Boston University Hariri Institute for Computing
Study Sponsor Location:
United States

Research Partner

Name of Partner Institution:
Innovations for Poverty Action Malawi
Type of Organization:

Intervention Overview


Firms will be randomized into one of three experimental arms:

  1. A control group (T0). The firm will receive an order without formal contracts. All the instruction will be given orally.
  2. A treatment group (T1) where the firm will receive an order with a written formal contract. In this arm, the print shop will receive exactly the same order as in the control group. However, the enumerator will give a written contract to the employee. The contract will specify the number of copies to be made, a description of the expected quality, a deadline (7 days after the order), and the price to be paid by the enumerator if the contract is respected. If the print shop accepts the task and accepts to sign the contract, the enumerator will come back 7 days later to collect the copies. The remainder of the intervention sequence (order assessment, etc.) will remain the same as has been described for the control group.
  3. A treatment group (T2) where the firm will receive an order with a written formal contract specifying a shorter deadline (2 days) for completion. In addition, the enumerator will informally inform the shop owner that the deadline is not a binding constraint as long as the quality of the realized order is adequate.
Theory of Change:

In many low- and middle-income countries, transactions between customers (buyers) and firms (sellers) are often conducted informally. In these contexts, receipts or other documentation are often not provided when purchasing goods, negotiating prices, or disputing the receipt or quality of purchases. Instead, informal contracts that are based on reputation and relationships between buyers and sellers serve to mitigate unforeseen contingencies, limit uncertainty on transaction outcomes, settle disputes over transactions, and improve performance standards and accountability. Though formal written agreements in these contexts could improve transparency and efficiency, the costs associated with their implementation and enforcement may not outweigh their benefits. In addition, formal contracts may not be used in contexts where they may crowd out informal mechanisms that are based on relationships and reputation.

While a large theoretical literature has identified mechanisms underlying the use of informal contracts, the role of informal contracts in dynamic settings, and the conceptual interactions between formal and informal contracts, many of the predictions that have been generated from this research have yet to be empirically tested. Fortunately, the empirical evidence on the use of informal contracts has grown in recent years. Findings from recent studies have shown that informal contracts may affect performance and productivity, and studies on enforcement costs have also, to various degrees, shown how these costs may affect performance.

Multiple Treatment Arms Evaluated?

Implementing Agency

Name of Organization:
Innovations for Poverty Action Malawi
Type of Organization:
NGO (International)

Program Funder

Name of Organization:
Boston University Hariri Institute for Computing
Type of Organization:
Research Institution/University

Intervention Timing

Intervention or Program Started at time of Registration?
Start Date:
End Date:
Evaluation Method

Evaluation Method Overview

Primary (or First) Evaluation Method:
Randomized control trial
Other (not Listed) Method:
Additional Evaluation Method (If Any):
Other (not Listed) Method:

Method Details

Details of Evaluation Approach:

Our main econometric specifications will 1) estimate the intent-to-treat (ITT) effect of being assigned a contract on the quality of the completed task; and 2) estimate the ITT effect of being assigned a tight contract compared to a loose contract. We will conduct several sub-group analyses in order to examine how our contracting intervention effects vary across key subpopulations. Subgroups of interest include: firms facing a high level of competition, small firms, firms that are innovative, firms that are owned by a family, firms which has a large share of non-permanent employees. Finally, robustness checks (5 percent and 10 percent sample truncations, coarsening of independent variables) and falsification tests, which include placebo regression, simulation, and resampling methods, will be conducted to ascertain the strength and significance of our estimates.

The following econometric specification will be used for estimating the intent-to-treat (ITT) effect of implementing the randomized design (formal written contract vs. informal oral agreement; order number) as follows:


Outcomes (Endpoints):

Outcome Measurement

By testing whether performance increases with the number of interactions between the client (the enumerator) and the print shop and randomizing the type of contract (formal written vs. verbal / informal) for each order, we will be able identify the relative importance of formal contracts and their complementarity / substitutability in a dynamic setting. We will measure performance on outcomes related to the quality of the order such as: the presence of errors in a completed printing order, whether the order was successfully completed, timeliness in completion of the order, and other metrics of quality, service readiness, and level of effort. Outcomes related to shop behavior and responses to being presented with a range of contract types will also be collected.

For each order of a given complexity that was submitted to a print shop, we will construct a series of variables that aim to capture order quality as follows:

  1. Did the shop accept the task?
  2. Did the shop deliver the task before the (informal) deadline?
  3. Did the shop deliver the task before the formal deadline (if any)?
  4. How many copies did the shop deliver? The correct amount?
  5. Number of copies that were correctly printed by the print shop.

For each task, we will check the number of copies that were correctly completed by the print shop. We will then create indexes of order quality for each order that a shop received.

Unit of Analysis:
print order

This experiment serves to test two hypothesis. First, we want to understand whether formal contracts have an impact on quality in the printing industry in Lilongwe. Second, we want to better understand the conditions under which formal contracts may be useful to increase output quality.

Unit of Intervention or Assignment:
print order
Number of Clusters in Sample:
100 print shops
Number of Individuals in Sample:
4-6 print orders per shop, total of up to 600 orders
Size of Treatment, Control, or Comparison Subsamples:
up to 240 orders for control, up to 180 orders for loose contract group (T1), up to 180 orders for tight contract group (T2)

Supplementary Files

Analysis Plan:
Other Documents:

Outcomes Data

This study is a three-armed randomized controlled trial that consists of a baseline survey followed by implementation of our print order intervention immediately following completion of the baseline. For each print order, an order initiation form and order completion form will be collected. A follow-up survey will be conducted one week after completion of the final order, at the time when the order will be picked up at the print firm.
Data Already Collected?
Data Previously Used?
Data Access:
Data Obtained by the Study Researchers?
Data Approval Process:
Approval Status:

Treatment Assignment Data

Participation or Assignment Information:
Data Obtained by the Study Researchers?
Data Previously Used?
Data Access:
Data Obtained by the Study Researchers?
Data Approval Process:
Approval Status:

Data Analysis

Data Analysis Status:

Study Materials

Upload Study Materials:
Approved Expedited Study Protocol: Pilot_2020_IRB_Expedited_Application_MVK_2-16-21.pdf

Registration Category

Registration Category:
Prospective, Category 1: Data for measuring impacts have not been collected

Completion Overview

Intervention Completion Date:
Data Collection Completion Date:
Unit of Analysis:
Clusters in Final Sample:
Total Observations in Final Sample:
Size of Treatment, Control, or Comparison Subsamples:


Preliminary Report:
Preliminary Report URL:
Summary of Findings:
Paper Summary:
Paper Citation:

Data Availability

Data Availability (Primary Data):
Date of Data Availability:
Data URL or Contact:
Access procedure:

Other Materials

Survey Instrument Links or Contact:
Program Files:
Program Files Links or Contact:
External Link:
External Link Description:
Description of Changes:

Study Stopped