There are many views about how a country develops. Some view institutions as the key determinant, while others emphasize the fundamental importance of human capital. Still others highlight the importance of infrastructure, while the World Bank and other international organizations have argued for improving the overall business environment in which firms operate. Finally, a recent strand of literature has emphasized the importance of agglomeration economies as a source of long-term growth. What, however, are the relative explanatory power of these alternative, though not necessarily-mutually exclusive, views? And are their effects specific to the context such as the level of development, the sector in which a firm operates, firm size and age? Answering these questions is important because governments only have limited resources to deal with key challenges. If there are bottlenecks to a country’s development, it is important to diagnose these to provide a sounder basis for policy.
In recent decades, there has been extensive cross-country literature that seeks to explore the determinants of long-run growth. However, a consensus has emerged that it is hard to distinguish between different determinants using cross-country data alone. Thus, a few researchers have tried to rely on sub-national data to address the afore-mentioned research questions. To overcome the difficulties associated with purely cross-country data and better distinguish between different potential determinants of growth, Reyes, Roberts and I (2017) relied on sub-national variations across countries.
We draw on World Bank Enterprise Survey (WBES) data for a sample which consists of up to 80,000 firm-year observations covering 709 cities in 128 different countries. We use this dataset to investigate how variations in the business environment (BE) across cities within countries are associated with both firm employment and productivity growth. In doing so, we adopt a broad category of the BE which encompasses many of the alternative explanations of long-run development that were mentioned at the outset of the blog.
The BE includes the following three large categories. First, we examine Basic BE, that is, those aspects often viewed as basic and fundamental for development. It includes basic functions of government protection, which include containing corruption and providing basic safety (“government protection”). It also includes a good supply of human capital and the availability of infrastructure (e.g. power supply and internet technology), as well as access to finance, which is widely viewed as a key BE element by a large swath of literature. Second, we investigate the Refined BE, which includes, inter alia, the barriers to entry and exit (such as access to land) and labor regulations that firms face, as well as the tax environment within which firms operate. The third, and final, dimension is the Agglomeration Environment. This incorporates whether a firm is in a large city, defined here as a city which is either a national capital or which has in excess of a million residents. It also includes what we refer to as “capacity agglomeration,” which refers to the extent to which a firm is joined by “capable” firms (i.e., the share of firms that employ more than 50 employees).
The firms within the WBES data that we use are randomly selected and are representative of the non-agricultural private sector in each economy. Unlike purely cross-country studies, the size of our sample is such that we can more confidently disentangle the role of the different elements of the BE in driving firm growth. Unlike the Doing Business (DB) project in which the indicators are de jure measures based on the answers of a few mid-sized firms (with the same firm size across the world), the BE measures in our data set are de facto, measuring actual experiences of firms. Since enforcement is a big issue in developing countries, de facto BE measures are likely to be more relevant for understanding the relationship between the BE and firm performance.
Our empirical investigation yields many new findings. First, Basic BE - such as the availability of modern infrastructure (internet), access to finance, and government protection - and the agglomeration environment are especially important BE determinants for firm growth. Second, we document evidence of significant heterogeneity of BE effects across different types of firms. To begin with, the environment which is amenable to the growth of high-growth firms is quite different to that for other firms. While, for the latter group, basic government protection and capacity agglomeration are important, high-growth firms benefit relatively more from modern infrastructure, access to finance, lower Land Access Obstacle and lower Tax Burden Obstacle, and from being in big cities. High-productivity-growth firms benefit significantly more from the trade credit network.
We also find that the BE effects depend on firm size, age and sector. With a few exceptions, small firms tend to need a stronger Basic and Refined BE than large and medium enterprises (LMEs). For example, in terms of effects on employment growth, LMEs benefit relatively more from the existence of labor flexibility and a strong environment in terms of both linkages between firms (e.g., access to trade credit finance) and agglomeration. Furthermore, mature firms benefit relatively more when corruption is contained, while young firms benefit more from modern infrastructure and Refined BE (i.e., ease in entry, labor flexibility). Interestingly, relative to those in manufacturing, firms in the service sector are much more sensitive to most elements of the BE. An exception is that manufacturing firms benefit relatively more from access to bank finance.
A novel finding is that the key ingredients for a good BE also differ according to the level of development of the country in which a firm is located. Access to bank financing and capacity agglomeration are more important for facilitating firm productivity growth in lower-income countries. Interestingly, trade credit access plays a positive role only in middle and lower-middle-income countries. However, most BE elements prove to be income-invariant in terms of facilitating both employment and labor productivity growth.
Our research yields several key implications. First, it is important to consider heterogeneous effects of the BE. Small firms (vs. LMEs) and services firms (vs. manufacturing firms) tend to be more sensitive to the soundness of the Basic BE and the Refined BE. Larger firms and high-growth firms tend to require stronger labor flexibility. Manufacturing firms (vs. service firms) tend to benefit more from access to bank financing. Small firms (vs. LMEs) and low-growth firms benefit more from a good agglomeration environment. Young firms (vs. old firms) benefit more from modern infrastructure and Other Refined BE (i.e., ease in entry, labor flexibility). For lower-income countries, access to bank financing and capacity agglomeration are more important for facilitating firm labor productivity growth.
Second, though we observe important heterogeneity of the BE effects, we do find several key elements to be critically important in most contexts: access to bank finance, modern infrastructure, and agglomeration (especially capacity agglomeration and being in a big city). Thus, it is important to add the agglomeration environment to the list of key BE ingredients. This category has emerged as a key determinant of firm growth across countries, perhaps as important as many of the traditional BE elements that we emphasize such as human capital, institutions, access to finance, and so on; moreover, it is a key element that facilitates productivity growth in low-income countries, thus allowing them to catch up.