Looking at the literature on informality, one thing that stands out is the small size of the informal firms. In fact, firm-size is one of the criteria used by ILO  and individual researchers to draw the line between formal and informal firms. Many informal firms, however defined, are operated by the owner herself or himself and without any other employees, with few having more than five employees.
The small size of informal firms is often regarded as one of the reasons why informal firms are not as efficient and dynamic as the formal sector firms. One thing that crossed my mind while glancing through the literature is that there is actually very little research on the relevance of firm-size within the informal sector. Which leads me to ponder, does firm-size really matter within the informal sector? Contrast this with the voluminous literature on how firm-size matters for the formal sector firms – in terms of exporting, R&D, access to finance, etc.
Fortunately, the Enterprise Surveys  datasets are rich with surveys set in Africa and Latin America and the Caribbean on informal firms. I started looking into these surveys (for 10 African countries) with the prior assumption that since firm-size drives a large part of formal-informal distinction, informal firms should look more like the formal sector firms as they increase in size. That is, compared with the small informal firms, large informal firms should be closer to the formal sector firms in terms of productivity, access to finance, etc. In fact, there are already some studies  that show that this is indeed true of firm productivity.
As it turned out, my prior was not entirely accurate (details here ). True, I found some areas such as use of bank accounts and loans from external sources where the prior was correct. There were some areas such as access to infrastructure (quality of power and water supply) where small and large informal firms showed no significant difference. More surprisingly, for the two performance indicators available – output per worker and employment growth rate, I found completely opposite results to my prior. That is, small informal firms had much higher output per worker than the large informal firms, and this difference is robust to a large number of controls for regional and firm characteristics (see the figure below). The same holds for employment growth.
Despite its large presence in many parts of the world, not much is known about the structure of the informal sector. Earlier research focused on the distinction between formal and informal sectors. Later research did explore heterogeneity within the informal sector such as the difference between successful vs. struggling informal businesses and between entrepreneurs who are in the informal sector because of lack of alternative formal employment opportunities vs. those that see opportunity and better income generation in the informal sector. However, as the results discussed above suggest, much remains to be done to better understand the relevance of firm-size, age, location, etc., for firms within the informal sector. I hope that the post draws attention to this important gap in the literature and that we see more research in this area in the near future.
Note: The figure is a partial scatter plot obtained after controlling for region fixed effects. The X-axis measures (log of) number of employees at the firm in a normal month over the last year and the Y-axis equals (log of) output (in USD) per worker in a normal month over the last year. The figure is based on data for 530 informal firms in seven countries in Africa for which data are available.