Better Together? Returns to Capital, Group Consulting, and Inter-firm Relationships Among Small Firms in Ghana

Last registered on January 13, 2025

Pre-Trial

Trial Information

General Information

Title
Better Together? Returns to Capital, Group Consulting, and Inter-firm Relationships Among Small Firms in Ghana
RCT ID
AEARCTR-0014260
Initial registration date
January 13, 2025

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
January 13, 2025, 2:11 PM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Mount Holyoke College

Other Primary Investigator(s)

PI Affiliation
Columbia Business School
PI Affiliation
World Bank

Additional Trial Information

Status
On going
Start date
2023-07-01
End date
2027-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Entrepreneurs in developing countries often have small and stagnant businesses, frequently reporting capital constraints and failing to implement standard best management practices. There is evidence that providing capital alone spurs growth while providing consulting services (group or individual) also improves management practices for such firms. In the case of group consulting, firm growth also increased significantly, delivering a cheaper yet superior alternative to individual consulting (De Mel et al. 2008; McKenzie & Woodruff 2008; Fafchamps et al. 2014; McKenzie, 2017; Iacovone, Maloney, and McKenzie, 2022). However, the mechanisms underlying the effectiveness of group consulting remain unknown, and any complementarities from jointly providing it with capital among the same firms is still an open question. We extend this literature, incorporating promising evidence on peer learning among entrepreneurs from Cai and Szeidl (2018), by evaluating the impact of providing 1,000 small firms (6-30 employees) in Ghana with scale-up grants, group-based managerial consulting, and peer learning opportunities in a randomized control trial (RCT). Specifically, we design our research to examine mechanisms underlying the effectiveness of group consulting and test for complementarities between large grants and group consulting on revenue and employment in a scalable setting. In a linked survey experiment, we also explore the possibility of eliciting information about participating entrepreneurs’ marginal returns (MRs) to the business support interventions and future business performance from their business networks of peer entrepreneurs, suppliers, and customers.
External Link(s)

Registration Citation

Citation
Awadey, Amanda, Laura Boudreau and Elwyn Davies. 2025. "Better Together? Returns to Capital, Group Consulting, and Inter-firm Relationships Among Small Firms in Ghana." AEA RCT Registry. January 13. https://doi.org/10.1257/rct.14260-1.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
We conduct an RCT to evaluate the impact of a large-scale business support program, the Enterprise Grow Program, implemented by the Government of Ghana through the Ghana Enterprises Agency (GEA) in partnership with the World Bank. In the RCT, 1,000 small enterprises (sized 6-30 employees) are randomly assigned to receive large grants of variable size (estimated mean: $10,000) and/or group-based managerial consulting (Iacovone, Maloney & McKenzie, 2022), group-based peer learning, and control conditions. Using this research design, we identify the causal effect of the program’s interventions and explore the underlying mechanisms of the effectiveness of group consulting and complementarities with grants in a scalable setting. Peer learning can be a cheaper alternative to group consulting and of policy relevance. We also examine heterogeneity in the marginal returns to capital, managerial group consulting, and peer networks/learning across small firms. With this, we can provide policy-relevant evidence on how these types of business support services could be targeted by firm type.

In a related prediction survey experiment (AEA’s RCT registry (AEARCTR-0013786)), we leverage this RCT to examine the ability of entrepreneurs’ business networks – their peers, suppliers, and customers – to predict entrepreneurs’ marginal returns to the program’s interventions and their future business performance in the short to longer term. By combining these experiments, we can obtain unbiased estimates of the marginal returns to these interventions across the distribution of predicted returns, allowing us to test the accuracy of marginal return predictions, and to measure the predictive power of information gathered about entrepreneurs’ absolute and relative performance.

This RCT includes firms from the GEA’s four calls for applications rolled out over two years. The program aims to use a rigorous screening process to target “high-growth potential” small and medium-sized enterprises (SMEs) – SMEs with promising prospects of scaling up their operations through increasing sales and job creation. Our study’s sample focuses on 1,000 small firms (6-30 employees) in the program. The treatment arms are: group-based consulting only, group-based peer learning only, group-based consulting and cash grant, cash grant only and a control group.

By combining the group-based consulting treatment and the group-based peer learning treatment in the same RCT, we can evaluate their relative performance and begin to disentangle possible underlying group mechanisms at play using experimental games. Furthermore, our research design allows us to examine heterogeneity in the marginal returns to capital, managerial group consulting, and peer networks/learning across small firms along many characteristics.
Intervention Start Date
2023-08-16
Intervention End Date
2025-06-30

Primary Outcomes

Primary Outcomes (end points)
Primary outcomes are revenue and employment
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary outcomes are management practices, investment levels, adoption of new technologies, number of new customers and suppliers, exporting products, cost structure and profits.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The study is designed as a randomized controlled trial (RCT), with eligible firms being randomized across five treatment arms: (1) control (no grant, only 360 assessment), (2) peer learning without grant, (3) group consulting without grant, (4) grant with only 360 assessment, (5) group consulting with grant. Each treatment arm has a target of 200 firms to be included.

Across all four cohorts, firms apply in a two-stage process. In the first stage, firms apply through an online portal. Based on the application data, an automated score is generated, and about 2,000 top-scoring firms are selected. Selected firms are invited to participate in a 4-day business management training.

In the second stage, each participant from the business management training is invited to submit a business proposal highlighting areas for improvement and goals they expect to reach during the program. These proposals are screened and assessed by technical experts based on technical criteria (e.g., feasibility or potential). The successful firms (about 1000 across the four cohorts) are then eligible for randomization by the research team into five treatment arms.
Experimental Design Details
Not available
Randomization Method
Randomization was done by a computer, using Stata DO-files. Each randomization has a seed drawn from random.org (using “set seed”), with a number between 0 and 10^9.
Randomization Unit
Firm
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
For most firms, individual randomization was used (with strata applied for region, sector and networking session participation). Cluster randomization is only applied in regions with fewer than 6 firms. This accounts for an estimated 5-6 percent of firms (in cohort 1, 17 out of 331 firms; in cohort 2, 13 out of 211 firms).
Sample size: planned number of observations
950 firms individually and 50 firms in clusters
Sample size (or number of clusters) by treatment arms
Each treatment arm will have approximately 200 firms.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We conducted power calculations using the “pcpanel” package in Stata, which allows one to perform power calculations by simulation using an existing dataset. We used the first stage of the 2014 Ghana Integrated Business Survey (IBES), which is an economic census conducted on the full set of economic units across all sectors of the Ghanaian economy. The IBES includes four measures of firms’ employment taken over nine months. While the research team aims to measure firms’ outcomes over a longer period, we are confident that we can follow firms for at least one year. Hence, we believe these data allow us to base our analysis on realistic assumptions. In the calculations, we focused on employment, as detailed information on revenue were not available. We apply the Bonferroni adjustment for 2 primary outcomes with 4 hypothesis tests each, which sets α = 0.0125. Under the following assumptions: time fixed effects, a control for the baseline value of the dependent variable, three post-treatment observations, n=200 per treatment arm, and an evenly split sample between treatment and control, we are able to detect an effect size of 15% of a standard deviation in employment with 96% power. This is a difference of 0.9 employees on an average firm size of 11.4 (an 8% change). We think this is an acceptable level of precision.
IRB

Institutional Review Boards (IRBs)

IRB Name
Columbia University Administrative Review Committee
IRB Approval Date
2022-02-10
IRB Approval Number
IRB-AAAT9400