SYSTEMIC RISKS AND FINANCIAL FRAGILITY IN A SMALL OPEN ECONOMY: THE CASE OF BOSNIA-HERZEGOVINA

Abstract
This thesis investigates the sources of financial system fragility in a small open economy with a
traditional banking system, with a focus on Bosnia and Herzegovina. Conducting research on Western
Balkan countries is challenging given the shortness of time series, unrepresentative samples,
numerous structural breaks, poor quality data and the historical absence of the phenomenon that is
the focus of investigation. For this reason, the common assumption that the findings associated
with other regions or countries are applicable to the Western Balkans is rejected. Instead, two
measures of systemic risk are constructed to assess a country’s financial system fragility that
reflects the specific characteristics of a country such as Bosnia and Herzegovina. The liquidity
index measures how vulnerable the financial system is to a currency crisis represented by the
abandonment of the currency board arrangement. The solvency index is an indicator of banking system
fragility at a point in time. Changes over time in both these measures of systemic risk are related
to changes in a set of macroeconomic and banking sector specific variables. This thesis contributes
to a better understanding of financial system fragility in Bosnia and Herzegovina and similar
countries in several ways. It is found that both country and period specifics must be accounted
for. Accordingly, each country should develop its own tailored measure of systemic risk, since some
of the widely used set of indicators, such as the level of foreign reserves, may distort the
perception of risk. A disaggregated approach to systemic analysis is favoured: it is more efficient
to interpret two measures of systemic risk jointly rather than to merge them into a single
indicator. However, there are substantial gains in modelling the risks of banking and currency
crises as a system. It is demonstrated that even in a country with a simple financial system and
dominant banking sector a single model cannot explain the evolution of systemic risk over the
cycle. The nature of the risk factors, their relations with the perceived level of fragility, as
well as the relationship between the measures of systemic risk were found to differ in pre-shock
from the post-shock periods. Finally, it is shown that even simple financial systems are inherently
unstable, with destabilizing relationships between the risks of banking and currency crises and
developments in the real economy. It is concluded that developing a set of country-specific risk
measures that indicate the evolution of the risk of banking or currency crises is an imperative.