Response of angel investors to information on emission reduction potential of start-ups

How to cite

Brahm, Francisco; Teng, Nina; Li, Xia; Rizzi, Claudio. 2024. "Response of angel investors to information on emission reduction potential of start-ups." AEA RCT Registry. AEARCTR-0014943. https://doi.org/10.1257/rct.14943-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Francisco Brahm* , Nina Teng , Xia Li , Claudio Rizzi

* Primary investigator

Registry ID
AEARCTR-0014943
DOI
https://doi.org/10.1257/rct.14943-1.0
First registered
Last updated
Status
In development
Trial dates
2024-12-13 – 2025-06-30
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/14943

Abstract

In this project we will study whether the emission reduction potential (ERP) of climatech start-ups affects investors' behaviour and financial decisions. Recently, reliable and fine grained ERP measures having been created by third parties that allow to have an estimation of the amount of carbon emissions that a particular start-up may be able to reduce (as compared to the state of the art). The climatech sector has received much attention and funds, but no evidence exist to date about whether ERP affects start-ups financial success in funding stages or latter outcomes (e.g., IPO). The project as a whole involves a regression analysis using Pitchbook data (a database tracking startups and investors), and two field experiments. In this pre-registration we are preregistering the second experiment.

The setting of our field experiment is a set of 21 early-stage climatech start-ups that will be evaluated by 25 angel investors (that are part of an angel investor network in South East Asia). The outcomes of interest (our dependent variables) are the evaluation score and the investment interest as expressed by the 25 angel investors. Our randomization of start-ups will be done "within investors", meaning that for each investor we will randomly select start-ups to be in the control or one of two treatment conditions. The randomization will be performed on the standardized template that evaluators will receive that describe the startups. This template as the usual information that early stage investor evaluate, such as description of the founding team, description of the idea, assessment of potential market, etc. One section of this template will have information on ERP, describing what ERP is, and the source of the ERP information we use. We will experimentally manipulate the ERP information in this section. The control group will indicate that the ERP information is "not available for this start-up". The first treatment condition will indicate that "This start-up is below the median ERP across all the start-ups evaluated in this program". The second treatment will indicate that "This start-up is above the median ERP across all the start-ups evaluated in this program". The comparison of the two treatment and the control will indicate whether the presence of ERP information affects investors' evaluation and selection. The comparison between the two treatments will indicate whether the level of reductions in emissions affects investors' evaluation and selection.

Experimental design

The setting of our field experiment is a set of 21 early-stage climatech start-ups that will be evaluated by 25 angel investors (that are part of an angel investor network in South East Asia). The outcomes of interest (our dependent variables) are the evaluation score and the investment interest as expressed by the 25 angel investors. Our randomization of start-ups will be done "within investors", meaning that for each investor we will randomly select start-ups to be in the control or one of two treatment conditions. The randomization will be performed on the standardized template that evaluators will receive that describe the startups. This template as the usual information that early stage investor evaluate, such as description of the founding team, description of the idea, assessment of potential market, etc. One section of this template will have information on Emission Reductions Potential (ERP), describing what ERP is, and the source of the ERP information we use. We will experimentally manipulate the ERP information in this section. The control group will indicate that the ERP information is "not available for this start-up". The first treatment condition will indicate that "This start-up is below the median ERP across all the start-ups evaluated in this program". The second treatment will indicate that "This start-up is above the median ERP across all the start-ups evaluated in this program". The comparison of the two treatment and the control will indicate whether the presence of ERP information affects investors' evaluation and selection. The comparison between the two treatments will indicate whether the level of reductions in emissions affects investors' evaluation and selection.

Design details

Randomization method
The experiment will be performed via qualtrics. We have coded in qualtrics the randomization of the treatments as follow: for each evaluator, we will randomly select 7 startup to be in the control group, 7 to be in the low ERP treatment, and 7 to be in the high ERP treatment.
Randomization unit
Within evaluator, the randomization unit is the start-up. But given that across evaluators a start-up will end up in expectation 1/3 in control, 1/3 in low ERP treatment and 1/3 in high ERP treatment, the "real" randomization unit is the startup-evaluator pair. We have 21 x 25 = 525 pairs. In order to avoid "order effects" in the evaluation, we will randomize the order in which start-ups will be evaluated by evaluators.
Clusters
We don't randomize at the cluster level. However, evaluators are clusters, thus we will cluster our standard errors at the evaluator level.
Observations
We have 21 startups that will be ebaluated by 25 investors, and thus we will have 21 x 25 = 525 observations.
Arms
Control = 175 start-up/evaluator pairs High ERP Treatment = 175 start-up/evaluator pairs Low ERP Treatment = 175 start-up/evaluator pairs
Primary outcomes
We will have four likert question (1 to 5), with 1 "definitively not", 2 "probably not", 3 "Might or Might not", 4 "Probably yes", 5 "Definitively yes": Would your organisation invest in this startup? Would you invest in this startup? Do you believe that this startup will be a commercial success? Would you like to connect to this startup?

Cited by

No citation data available for this registration yet.

Keywords