For many, the main appeal of self-employment is being your own boss. The freedom to decide how to reach your goals, without working under a watchful eye supervising every move, is certainly an enticing prospect. After all, isn't professional independence demonstrated by not having to report to others about what one is doing?
However, alongside independence entrepreneurs need to perform. Performance is the consequence of commitment and effort towards a given goal. One key benefit of a supervisor is their role in prompting commitment and effort through encouragement, social pressure and incentives. Given that dependence plays an important role in performance, what if professional independence actually impedes the performance-enhancing benefits of having to report progress to a third party? Do entrepreneurs perform differently when faced with an exogenously imposed accountability structure? Or perhaps entrepreneurs, unlike employees, are so driven by internal forces that they do not need external mechanisms to push them forward?
We aim to explore this tension between freedom and dependence, because it is not clear whether it suits entrepreneurs to follow their desire for professional independence, or whether relinquishing some freedom may actually help entrepreneurs reach their startup’s performance goals.
This issue is important for both entrepreneurs and policymakers. Let's assume for a minute that a bias towards independence inhibits entrepreneurs from reaching their performance goals and that accountability structures drive entrepreneurial performance. If this were the case, entrepreneurs may actually benefit from strategically self-regulating their wish for independence. Thus, the process of developing accountability structures (like an advisory board, for example), would keep entrepreneurs on a high-performance trajectory. In the case of policymakers, entrepreneurship promotion programmes (such as business accelerators) may actually benefit from investing in, and setting up, processes aimed at providing greater supervision over entrepreneurs.
In contrast, if our research were to find that accountability structures have no effect (or even a negative effect) on performance, this project would also provide valuable evidence to entrepreneurs and policymakers about the ineffectiveness (or even harmfulness) of supervisory structures.
What is structured accountability?
Structured accountability encompasses the actions of being encouraged to reflect on how one should reach ones goals; a guided discussion about these goals with others; a request to commit to the tasks to reach these goals; and a review of the success of previously committed tasks. It is believed that by creating a structure that encourages entrepreneurs to systematically and periodically go through this cycle to become accountable for their self-devised plans would, according to existing rationale, lead to more favorable outcomes in terms of new venture performance.
What we’re going to do
To test this rationale, we partnered with a major business accelerator and designed an accountability structure by which participants periodically (roughly every four weeks) meet with a group of (around eight) peers and an accelerator executive. At each one of these meetings, participants are asked to reflect on the success of the tasks they committed to during the previous gathering, and share the tasks they plan to execute before the next gathering.
In order to test our hypothesis on the impact of structured accountability, we set up a simple two-arm randomised controlled trial within our partnering business accelerator. Participants are randomly assigned either to the treatment or control group. All participants are exposed to periodic, group gatherings. However, only treated participants are subject to structured accountability. After the six-month intervention, we gather four measures of startup performance (survival, jobs created, capital raised, and market traction), which we repeat six months later for a second wave of performance data. In addition to the quantitative analysis, we plan to conduct in-depth interviews to support our findings.
The trial will run through four accelerator cohorts, ending in January of 2019, and we expect to have preliminary results by mid 2019.