Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type



Business and Economics



First Advisor

Dr. Christopher Bollinger


As a post-Soviet economy, Ukraine has inherited substantial production assets and qualified personnel. However, the economy was dominated by large-scale enterprises designed for much bigger markets. After the collapse of the Soviet Union Ukrainian firms faced lack of planning, breaks in contacts with their former suppliers and customers, and distortion of prices. There was a clear need in restructuring of the entire economy. Restructuring included splitting firms into smaller parts and privatization. The first phase of transition was completed by 2000 when the output grew for the first time after a long recession in nineties, and most firms became private property.

In this work I explore trends in geographic and industrial concentration of Ukrainian manufacturing firms over the period of 2001 to 2005. I found that this period was characterized by relocation of firms between sectors and between regions, as well as by an increase in economic concentration of industries. The speed of adjustment was different for various sectors and even for different industries within manufacturing. Even though the economy is still dominated by large firms, the average firm size decreases due to a rapid growth in the number of new firms. Geographically, manufacturing tends to increasingly concentrate mostly around a few big cities, apparently at the expense of other regions.

I also estimate the external scale effects and compare them with Western studies. In particular I focus on machinery and high tech. I found strong localization and urbanization effects in both industry groups. An important contribution of this work is the analysis of the effect of ownership structure on agglomeration economies. I found that private firms tend to enjoy external scale effects to a greater extent than state owned, and foreign owned firms appear to be the most efficient in extracting benefits form agglomeration.

Aggregation of the data may distort the estimates of agglomeration effects. I show that most effects take place at the nearest neighborhoods. When the physical distance between firms increases agglomeration effects attenuate quickly. However, localization effects reveal themselves at different level of industrial aggregation for various industries. This may reflect more complicated relationships within sectors and requires further analysis.