top of page
Search
Small Firm Diaries Team

A New Framework for Understanding Small Firms

We launched the Small Firm Diaries because after decades of work trying to help microenterprises to grow into small firms, there was still very little work to describe and understand existing small firms.


Given the dearth of data on small firms we created a simple dividing line between micro and small: at least one paid worker. Our delineation between small and medium was a bit more nebulous: no full-time, non-owner managers (we used less than 20 paid workers as a fuzzy proxy for this measure). Now that we have 12 months of diary data on 794 firms in six countries, we are in a better position to talk about small firms in much more detail than our initial, arbitrary lines. 


With this blog post, we share a “discussion draft” of a framework for describing policy-relevant characteristics of the businesses that we call small firms. This framework reflects the elements that small firms generally share, while also teasing out important elements of heterogeneity among them.


We will be using this framework in our analysis, including in the firm profiles we publish in the coming months. We encourage others who want to move beyond thinking of micro, small, and medium businesses as an undifferentiated group to use it as well and share their input (and critiques and questions!). 


Where we started: Defining our sample population

To build our sample population when we started the Small Firm Diaries in 2021, we established a set of inclusion criteria that reflected our (and our partners’ and funders’) interests in poverty alleviation, sustainable growth, and women’s empowerment. 


The most important inclusion criterion, note above, related to jobs. We included only firms that have at least one and not more than 20 paid workers. That intentionally breaks with definitions established by international organizations like the IFC and the OECD, which classify microenterprises as having between zero and 10 workers. Yet research shows that hiring even one worker represents a significant inflection point for a business. Firms qualified as small in our definition, therefore, if they paid at least one non-family member.


That still includes a lot of different firms. Additional criteria included industry (we included firms involved in agri-processing, light manufacturing, and some services), and location (we sought out firms in low-income neighborhoods, and looked for clusters of firms mainly in urban and peri-urban areas). We also set a goal of having at least 30% women-owned firms.


A Look Inside Small Firms


Having now gathered a year’s worth of detailed data on the firms captured with these criteria, we’re in a position to enrich our description of what makes a small firm beyond the number of workers. We think such a description is useful not only to carve out a well-defined space for them between “typical microenterprises” and bigger, more professionalized “medium” firms, but also to highlight some of the operational challenges and barriers to growth that small firms have in common, and to see axes on which they differ.


On this last point, even as we argue that small firms are a distinct group, there is significant (and policy-relevant) variation among them. Small firms are far from a monolithic category, and there is considerable diversity even in our constrained sample.


Introducing the Small Firm Framework


Our framework spans several broad themes—labor; capitalization and financial inclusion; formality; and volatility and growth. Across those themes we include eight different dimensions, which we think are useful to describe small firms, and the behaviors they exhibit. In some cases it’s clear that the firms on the far left of the spectrum for certain dimensions are more “like micro firms,” and firms that fall on the right side of the spectrum are more “like medium firms.” But for other metrics, it’s not so clear cut that position along the spectrum correlates with “micro vs. small vs. medium.”


FIGURE 1: The Small Firm Framework: Visualizing averages from the Kenyan sample


table with small firm framework dimensions

As shown in the summary chart above, we use three different metrics to cover the theme of labor. We start with the average number of workers that the firm pays per month. Firms across the Small Firm Diaries global sample have an average of two to three paid workers at any given time.

A simple count is not enough, however, to distinguish between firms that have casual workers from those with full-time workers with job security, or something in between (most likely in our population). To see these differences, the framework includes an index for worker retention, calculated by dividing the number of workers paid during eight or more months of the study by the total number of workers. To round out the picture on labor, we also capture the share of total expenses spent on labor, to reflect how much of a firm’s resources support jobs. Although this figure can be driven by industry (because certain goods and types of goods are essentially more labor-intensive to produce than others), we think it is an important dimension for understanding a firm and its possible growth trajectories.


The next four components of the framework relate to a firm’s level of capitalization and—relatedly—degree of financial inclusion. We look at the relative use of bank accounts and direct payments compared with cash to conduct business. We also analyze the use of credit on a continuum where we attempt to capture whether the firm is able to meet all of its credit needs through formal lenders (which typically offer larger amounts of credit at lower interest rates, but with stricter criteria), or whether—as is typically true of our small firm population—it holds a portfolio of loans from a variety of sources including informal loans from friends and family, informal lenders, MFIs (5), government, and banks.


The sixth metric reflects the degree to which revenues and expenses match closely over time. Classically, microenterprises run simple business models and rely on short-term transactions with very little spare capital on hand. These firms spend only when they earn, and earn only when they spend, thus the timing of expenses matters a great deal. On the other side of the spectrum are firms with healthy amounts of working capital, whether in the form of available cash, stocks of inventory, or raw materials. They can make large purchases or invest in future productivity, such as by hiring more workers or expanding operations, independent of revenue streams. For them, the timing of expenses matters much less than for micro firms. Most small firms occupy a middle ground, running business models with just enough complexity that their earnings and expenses are somewhat separate but not completely divorced from each other. 


The ability to make bulk purchases is another metric that stands as a proxy for whether a small firm has the capital resources to grow. Bulk purchases would allow a firm to, for instance, stock up on raw materials or inventory (and increase margins by buying at lower cost), or to make some other “lumpy” expenditure like a new piece of equipment.


The eighth metric relates to formality, which we interpreted broadly. Each firm owner described the different types of registrations they have, including with tax authorities, chambers of commerce, local neighborhood councils, and certifications, such as from industry boards or quality-control offices. We found that small firms frequently have some type of official registration, but often not the ones that they perceive to be too costly or cumbersome.


Finally, we turn to volatility and risk, and growth. To understand how much a firm’s revenue fluctuates from month to month, we use the CV of the firm’s median monthly revenue. In explaining the volatility both micro and small firms face, we’ve often used the metaphor of a raft tossed by ocean waves. Just as a larger boat would have the heft to remain relatively stable in powerful currents, larger firms would ideally rely on reserves of capital to guard against fluctuations in revenue or expenses. How much revenue volatility a firm faces is an imperfect but useful proxy for risk, specifically the likely amount of additional risk a firm may be willing to take to invest in growth. Growth often involves taking on additional risk—what if the investments in growth don’t pay off?—and a firm with predictable and stable revenues will, in general, be able to tolerate more additional risk.


In an environment with so much volatility it’s hard to see whether firms are growing from month to month or year to year. We found that most small firms are neither growing nor declining precipitously but rather doing what the captain of that small raft would do—hanging on tight, and working hard to inch forward despite the waves. We haven’t included a measure of growth in the framework (yet) because of how difficult it is to get a true picture of whether firms are growing or not, but in future analysis we’ll be sharing more about how we’re thinking about growth and our attempts to create a model that accurately estimates a firms’ growth trajectory by correcting for seasonality and inflation at the country and sector level. 


In the figure below, we put the framework in action to describe one of the firms in our sample.


Figure 2: A leatherwork firm in Kwale


Our first firm is a leather goods (primarily shoes and sandals) maker in Kwale, Kenya, owned by a man we’ll call James. 


This firm employs on average two workers per month (matching the average for the sample),. Over the course of the study, the firm recorded paying a total of five workers. None were paid during eight or more months, giving this firm the lowest possible worker retention rate (compared to the Kenyan sample average of 49%). Spending on labor amounted to 25% of total expenses, on par with the sample average. 

As it sells hand-made leather goods in a market in Kwale, this firm relies mainly on cash, conducting only 25% of transactions (by value) through electronic bank transfers or mobile money. This is lower than the average for Kenya by 20 percentage points. James reports neither having nor servicing loans of any kind during the study. The firm’s revenues and expenses are highly correlated at 85% (inverse correlation of 15%), meaning that the firm spends only when it earns, or earns only when it spends, and may be capital constrained. In line with this high correlation, the firm does not often accumulate large sums, as only 6% of purchases (by value) are bulk. James reports that he has registered his firm with the Kenyan tax authorities, meeting the highest bar for formality in the Kenyan context.


The CV of monthly revenue for this leatherwork business is .40, meaning that, on average, its monthly revenue tends to be 40% greater or lesser than its average monthly revenue. These large swings from month-to-month nonetheless are less than the sample average, of .53. 


Next steps

Over the coming weeks and months, as we continue to publish and disseminate firm profiles and policy briefs, you will see us using this framework to help quantify the circumstances of a firm or cluster of firms. 


As we use it, we may refine the metrics and make changes to how we calculate them. If you have ideas for other dimensions you think we should explore, reach out and let us know.

To start, check out our firm profile from Kenya: “On the Margins: Navigating the Fish Value Chain on Lake Victoria” (coming soon). 


Endnotes:

  1. Although our sampling criteria allowed for firms to have from 1 to 20 paid workers, most of the firms selected into the sample based on our comprehensive criteria list had less than 10 workers; on average they had between two and three workers.

  2. During data analysis we excluded the first 2 months of data for each firm because of the relatively higher likelihood of errors in the initial period of the study. An important feature of financial diaries research, characterized by frequent meetings repeated over a full year, is the familiarity and trust built between the research subject and researcher over time, which means that data quality increases as the weeks go by.

  3. It’s worth noting that small firms in Kenya offered workers more stable employment compared to most other countries we studied.

  4. The scale here is admittedly imperfect but the idea is to give a lower score to firms with loans that tend to be smaller, with higher interest rates, no collateral required, and depend on social risk for repayment, and higher scores to firms with loans that are larger, with lower interest rates, potentially some collateral required, and use traditional risk measurement.

  5. We find that microfinance institutions do not serve the small firm population in any significant way across most of the SFD countries. Kenya is an exception to this, where the same proportion of firms—23%—had loans from commercial banks as MFIs.

Recent Posts

See All

Comments


bottom of page