Networks, Information and Economic Volatility

This thesis makes a contribution to network theory and how it applies to economics. It consists of three self-contained papers. It predominantly considers how connections between economic entities can affect economic outcomes. In particular, the first two papers examine social learning, in the one case by applying it to portfolio choice, and in the other by conducting an experiment to determine how people incorporate information from others into their beliefs to achieve economic outcomes. The final paper looks at how sectoral shocks can be transmitted through the economy through the connections between industries.

The first paper, \textit{Social learning and financial markets: Can informed neighbors make up for a lack of financial acumen?}, allows financially- informed and uninformed agents to consult with each other on different social networks and examines what the benefits are to the two different groups of agents on the different networks in terms of financial investment.

The second paper, \textit{Incorporating information into beliefs on networks: An experiment}, studies how agents incorporate into their beliefs information they receive from other agents. In particular, it considers whether agents use the DeGroot model or a Bayesian approach. Further, it considers how the density of network connections affects the results of consultation.

The third paper, \textit{Sectoral shocks and aggregate volatility}, explores how sectoral shocks can propagate through the economy through the input-output networks. It devises a demand-side measure of industry influence and shows that such a demand-side measure is needed to fully understand the influence of industries with a case study of the Australian mining industry.