Can we predict a financial crisis
Prediction’s Essence VS Economics’ Research
Financial crisis is fluctuations and uncertainty of the society caused by economic, social and political deterioration. Numerous destructive effects, such as decrease in GDP, increase of unemployment and social panic, are caused by financial crises, so prediction of financial crisis has long been a major topic in Economics (Karacor 5). However, due to ineffective finance information processing and influence from unstable country’s economic state, timely prediction of financial crises accurately in a short term is impossible. To be specific, ineffective financial information processing completely based on past data disenables economists to predict financial crises in time, while instability of any country’s economy adds uncertainty to financial crises prediction throughout the world through financial contagion effect. Although some models are successfully used to predict some crises, they are vulnerable to randomness.
Uselessness of finance information processing solely based on past data stop economists from predicting financial crises in time. Signal approach, the most prevailing early warning system, is used to predict financial crisis and over 100 macroeconomic indicators are used by economists to analyze economic state, which can be traced back to 1998 (Kaminsky 6). However, it fails to predict dot-com bubbles and sub prime crisis, since it fails to predict the correlation between indicators and neglects the influence of new factors. On one hand, the correlations between different indicators before a crisis and in the normal state are different, while signal approach only calculates the relationship between indicators based on value range of indicators in the normal state (Kaminsky 10). On another hand, the influence of financial innovation cannot be predicted due to the deficiency of data. According to the National Bureau of Economic Research, financial innovation, such as collateralized debt obligations (CDOs) and credit default swaps (CDSs), is accused of causing financial crisis (Johnson 1). However, without previous data on the influence of them, economists is not able to predict financial crisis caused by them. Other prevailing models, such as cross country regression and MRD model, is based on the analysis of past data, so they cannot be used to predict a financial crisis too (Karacor 6). In addition, although volatility is commonly considered as one of the most important indicator of financial crisis, the correlation between volatility and financial crisis cannot be calculated due to deficient past information (Danielsson 3). Since economy is not free for anyone to conduct experiments to eliminate influence of randomness on past data or get more information about the financial crisis. Thus, complete dependence on past data, as a common basis of prediction, is not effective in the field of Economics. Moreover, influence from other countries will add more uncertainty to existed model for one country.
Instability of one country’s economy will influence the world econmy my through financial contagion effect. Due to different economic structure than developed countries, developing countries have lower economic stability, which indicates a higher risk of financial crisis. Developing countries economics has large economic dependence on one specific industry, such as oil for Turkey, and have debt problems (Avcı 10). In 1982, Latin American countries could not repay debt due to depreciation of their currency, which causes Sovereign debt crisis (Anderson 1). It shows that developing countries are easily influenced by external factors and harmed by the financial crisis. According to analysis of international financial crisis, financial contagion, the influence of one country’ economy to another country is an indispensable reason for most financial crises (Moser 2). Due to the global economy, countries’ economies states are closely related to one another through trade and investment. For instance, until 2018, China, as a developing country, has a large investment, which accounts for 20% of total investment, on the euro zone, including mostly developed countries (Canofari 3). As a result, the influence of developing countries on developed countries cannot be neglected. While developing countries suffered financial crises more possibly due to instability of economy, every country may be influenced by the instability of a certain country’s economy, which will add uncertainty to the national model for prediction. Additionally, even if researchers successfully build a model for one country, due to significant differences between countries, it cannot be applied to every country, which will add uncertainty to this model’s prediction through financial contagion effect. Consequently, influence between countries adds uncertainty to prediction of financial crisis. However, some model succeeded in predicting some crises.
Studies show that some modern models can predict several specific crises. Prediction of Turkish financial state with Artificial Neural Network based on past data give a good result, compared to the real situation, so it can be used in future prediction of financial crisis (Aydın1 41). Artificial Neural Network is famous for studying from the mistakes and changing the weight of input and the weight of small decisions made by part of the model to give several outputs. However, the correlation between macroeconomic indicators is established from data in the normal state to predict a crisis state. As a result, randomness of the data is still a problem, which decreases prediction accuracy. Also, 84 input of economic indicators to this model may not be enough to predict financial crises (Aydın1 38). As stated before, financial innovation’s influence won’t be predicted by this new model. New indicators and existed indicators, which are lack of further studies, always challenge the existed model. In conclusion, although financial information progressing with modern algorithm may be more useful than the old evaluation method, signal approach, the essence of prediction based on past data doesn’t change, so this model won’t successfully predict every crises.
Financial crises cause troubles in citizens’ life and economic structure of a country, desire to avoiding them encourages studies on prediction of financial crises. However, just like a drunk man who stumbles along the road, nobody knows where he will fall down or be stopped by another man, nobody can predict financial crisis due to uselessness of processing solely past financial data to build model and financial contagion between countries. Past data are not enough for economic analysis, which is proved by the inability of prediction based on it. Financial contagion from other countries intensifies the uncertainty of prediction of financial crises in one country. In future, the researchers should focus more on improving existed model to eliminate uncertainty, which can be used to predict cyclical recession. Nevertheless, for financial crises, how to solve it quickly when the tendency of the crisis exists is the most important.
Work Cited
Anderson, Spencer. “A History of the Past 40 Years in Financial Crises.” IFR.
Avcı, M. Ali and N. Oğuzhan Altay. “Predictability of Financial Crisis in Developing Countries: Turkey, Argentina and Thailand.”International Conference on Economic and Social Studies, (ICESoS’13), 10-11 May, 2013, Sarajevo.
Aydın1, Alev Dilek and Seyma Çalıskan Cavdar. “Prediction of Financial Crisis with Artificial Neural Network: An Empirical Analysis on Turkey.” International Journal of Financial Research Vol. 6, No.4, 2015.
Canofari, Paolo, and Alessandro Del Ponte. “Chinese and European Financial Systems: Instability Drivers and Contagion Channels.” International Advances in Economic Research, vol. 24, no. 4, 2018, pp. 311+. Questia School, www.questiaschool.com/read/1G1-563359699/chinese-and-european-financial-systems-instability. Accessed 2018.
Danielsson, Jon and Marcela Valenzuela. “Learning from History: Volatility and Financial Crises.” Finance and Economics Discussion Series, Divisions of Research & Statistics and Monetary Affairs, Federal Reserve Board, Washington, D.C.
Johnson, Simon and James Kwak. “Is Financial Innovation Good for the Economy.” May 2014.
Kaminsky, Graciela, Saul Lizondo and Carmen M. Reinhart. “Leading Indicators of Currency Crises.” International Monetary Fund, March, 1998.
Karacor, Zeynep and Korhan Gokmenoglu. “Predictability of Financial Crises: Testing K.R.L. Model in the Case f Turkey.”Annals of the Constantin Brâncuşi, University of Târgu Jiu, Economy Series, 2012.
Moser, Thomas. “What Is International Financial Contagion?” International Finance 6:2, 2003: pp. 157–1.