Studying the Network of Overnight Bank Loans in Italy
In Europe as in America, banks are required to maintain a certain amount of money as reserves that they must have on-hand at certain times throughout the year. In an attempt to prevent liquidity shocks, the European Central Bank requires that the reserves of any bank at the end of each month must be at least 2% of the total deposits and debts owned by the bank. In order to meet these requirements, banks often exchange liquidity with other banks in order to meet the reserve requirements. Large banks find it more profitable to lend out more money to businesses and pay smaller banks for loans to reach their liquidity reserves, while smaller banks may find it more profitable to simply loan out their excess reserves to big banks so those banks can meet their reserve requirements.
Studying the network of the overnight money market in Italy allows researchers to understand how the network is organized as well as predict any risks that may arise from the structure of the network. It was found that the network of Italian overnight interbank loans have a power law distribution of degrees. In other words, there are a high number of incredibly “popular” banks in the sense that they have exposure to a lot of other banks, and this “popularity” decreases exponentially. This power-law organization decreases the overall stability of the network and makes it incredibly susceptible to systemic failure. In the current organization, a small group of large banks borrow extensively from a large number of small creditors. This configuration presents problems as if one large bank is unable to pay back its loans, it could ripple throughout the network.
Researchers also looked at the evolution of the network over time to see if the network is becoming more or less susceptible to risk. They found while the majority of banks in 1999 had excess liquidity by the end of each month, in 2002 there was an overall shortage of liquidity and many frequent trades occurred at the end of the month drew closer. This infers that banks are more reliant on the complex network of overnight bank loans as time goes on.
After thoroughly studying the network of interconnections among Italian banks in the overnight market including creating an undirected graph connecting nodes of banks with edges if they have exposure to each other, the researchers conducting the experiment proposed policy solutions to minimize network risk. Because the overall connectedness of the network increases as the end of the month approaches, it increases systemic risk. One potential policy solution would be to encourage a schedule of reserve requirements that does require all banks to simultaneously meet their reserve requirements.
This study relates heavily to what is being studied in the Networks course. The “power law” description of popularity in the network is studied in order to understand the evolution of risk in the bank network. The long “tail” of this network shows that there are a small number of incredibly connected banks, which could spread failure across the network if any one of them were to fail. Additionally, this study dealt with the interconnectedness of the entire network and created an undirected graph where banks were represented by nodes. I particularly enjoyed this study as it recommended policies that the European Central Bank could make to decrease network risk.
Link to study: http://www.sciencedirect.com/science/article/pii/S0165188907000474