Friday, October 28, 2016

The inter-connected economy problem

While the twentieth century was all about globalization and an attempt to create a truly world economy, the twenty-first century seems to be going in the reverse direction. Various decisions like Brexit, Greece default etc. often questions the success of a globalized world.
Based on such thoughts, the following questions pop up:
  • How well connected are world economies?
  • What would happen if an economy falls?
  •  Should a country be worried about a possible default by a country with which does not have any trade relations?

While a decaying and troubling economies have numerous impact of the human life, my area of concern is primitively financial. It is essential to understand what is the impact of such failures of government on the value of currency for other countries.

As per Professor Gustav Cassel, of Sweden, the Stock Exchange has often been represented as an astonishingly sensitive barometer, which indicates beforehand what is going to happen in economic life.

Based on such an interpretation regarding the stock exchange I would like to measure the network that exists between the various stock exchanges of the world and to understand whether a domino effect exists within this network, i.e. if one of the exchanges fall, the other fall along with it as well.

To understand this network, we must understand the working of the stock exchange. A stock exchange refers to a market place where the equity shares and bonds of companies can be traded. The price of these shares is determined by the demand and supply of these securities. The demand and supply is determined by the profitability of the company (both current and future). With globalization, many companies are registered on multiple stock exchanges across borders. Further, many companies have procured investment or made investment in other countries. Also, many global companies act as a supplier or an intermediary to local businesses and therefore, there exist a mesh of international transactions within various companies.

Based on such understanding, we try to understand how a certain stock exchange can impact economies other than its home economy. This study cannot be done at an instance of time but has to be down over a period of time. For instance, if the study is done for a date which is prior to 2000, it could end with a result of China being a fringe economy, which in current timeline is not the case. Therefore, multiple events over the period needs to be consider to assess the impact of stock exchanges. Few of such events identified are as follows:
  • Brexit referendum
  • Greece default
  •  Lehman Brother bankruptcy
  •  Enron scandal
  •  Dot com burst
  • September’11 terrorist attack

Now, on to the trickiest part, that is creating a network between the various exchanges. All the exchange operates based on an index which measures the performance. If the companies listed are doing well, the index goes up otherwise, the index falls. To measure the impact of any event, we measure the change in the index of other countries with relation to the movement of the index of the home country. For instance, in event of the Brexit referendum, what was the degree of rise and fall in the world stock exchanges as compared to the London Stock exchange, was the change by the same degree, was there no impact or whether the index moved in the other direction.

An important factor to consider in this regard is the time zone and operating hours of the stock exchanges. Therefore, it would be essential not just to see the change in the movement on Day 0, that is the day on which the event happened but also the following 2-3 days which give an understanding on how the event is affecting the investment in the area. Another factor to consider is the market understanding of the scenario. With advent of technology and faster communication methods, analysts around the world make predictions about the outcomes of multiple events, while event like recession, terrorists act cannot be forecasted, analysts do predict on decisions on referendum etc. While it is difficult to determine when the analysts make these predictions and factor them into various investment being done, it can be said that an analysis of +- 7 days of the Day 0 is essential to understand the impact a certain event can have on the world economies.

While I have decided to take some additional time in putting this data together and defer my enrollment for the second module, I wish to social network analysis and machine learning techniques to create a system which can enable countries to identify their direct and in-direct dependency on various economy. This would allow them to better understand the forces impacting the global economy and take the necessary corrective steps.

1 comment:

Christopher Tunnard said...

This has good potential, but, as you say, the tricky part is defining what the network actually is, and, once you've done that, selecting the exchanges amongst and between which the network flows will be studied. No small task. Before you undertake this, have a good look on Google and the INSNA listserve, as there are several studies that have been done on networks within exchanges.

I hope you'll take this up someday.