Trade shocks and production networks in Uganda
The structure of firm networks matters. Given a certain network structure, it influences the amount of value created in an economy and determines an economy’s resilience to shocks. However, the structure is also likely to be endogenous to economic policy decisions. As such, it is difficult to disentangle the causal effect of a given network and the formation of that network.
There is still much to learn about what the economy-wide network looks like, especially in a developing economy, like Uganda. It is crucial to understand how shocks spread across firms through production networks, which should be the starting point for research on the propagation of shocks and network formation. The main reason for this knowledge gap is that detailed data on linkages between firms is rarely available, especially in developing countries. Uganda is an exception to this since it has detailed transaction-level data from the monthly Value-Added-Tax (VAT) declarations of firms. This provides us with a unique opportunity to study the network of firms in a developing country.
The focus of this project is two aspects of networks that are highly relevant to the general policy goals of the Ministry of Finance and Economic Development (MoFPED) and the Ugandan Revenue Authority (URA). Firstly, understanding when, and for what reasons, firms enter and exit the Ugandan economy. This has not been done before and is crucial to begin understanding how government policies can support the development of firms. Secondly, endogenise network formation, which will allow research on how the network corresponds to certain events. Also, research in entry and exit; this is an important starting point to gain a better understanding of how government policies affect the formation of firm networks in a developing country.
In order to achieve this, a firm network map of Uganda is constructed. Allowing comparison of the network of firms in Uganda, to a network in a developed economy. Then a dynamic analysis will be turned to. Given that the data ranges from 2012 to 2017, the evolution of networks over time can be assessed. After conducting these descriptive analyses, the effect of exogenous shocks on the network will be assessed. Finally, this study will endogenise to aspects that are generally assumed to be static in the literature.