Measuring Uncertainty in the Financial Sector: Insights and Challenges
The paper explores the concept of uncertainty in the financial sector, proposing a novel measure using expectation errors of managers. It discusses the impact of uncertainty on economic activity, drawing from relevant literature and financial services survey data. The study aims to address the lack of a proper measure of uncertainty in financial services, highlighting the importance of understanding and quantifying uncertainty for economic agents.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
MEASURING UNCERTAINTY IN THE FINANCIAL SECTOR AYTA ERDO AN Statistics Department EU Workshop on Recent Developments in Business and Consumer Surveys November 13-14, 2014 Brussels, Belgium * Corresponding author. aytac.erdogan@tcmb.gov.tr
Outline What is Uncertainty? Aim of the Paper Summary of Literature About Measuring Uncertainty Financial Services Survey Methodology Results Conclusion
What is Uncertainty? At a general level, uncertainty is defined as the conditional volatility of a disturbance that is unforecastable from the perspective of economic agents. A key concern for agents in the economy Higher uncertainty has adverse effects In spite of its importance, measuring uncertainty is a real challenge for statisticians.
Aim of the Paper Lack of a proper measure of uncertainty about financial services This study proposes a (proxy) measure, using expectation errors of managers in the financial sector The identifying assumption is that higher uncertainty is closely related with higher expectation errors. Relatively new survey, only visual evidence is found
Summary of Literature What is the impact of time-varying business uncertainty on economic activity Guiso and Parigi (1999) Bloom, Bond and van Reenen (2007) Bloom, Floetotto and Jaimovich (2009)
Summary of Literature Our work is closely following Bachmann, Elstner and Sims (2010) and Arslan et al. (2011) construct measures of time-varying uncertainty from business surveys. They examine their relationship with economic activity over the business cycle for German and Turkish data. Both studies use forecast errors of participants as a proxy for uncertainty, stemming from the identifying assumption that higher uncertainty causes higher forecasts errors. They both show that uncertainty has a strong relationship with output and can be used as a leading indicator Unlike them, we measure uncertainty in the financial sector by using Financial Services Survey data.
Financial Services Survey Financial Services Survey (FSS) is one of the surveys of The Joint Harmonized EU Programme of Business and Consumer Surveys. In Turkey, FSS was started in May 2012 and is fully harmonized with the EU programme. In Turkey, FSS covers 6 sectors Sector Number for Institutions Sector Share (%) Coverage (%) Banking 49 92.5 100.0 Insurance 20 3.3 85.6 Financial Leasing 11 1.6 88.3 Factoring 28 1.1 85.2 Financing 5 0.8 84.8 Brokerage 20 0.7 87.2
Survey Questions The questionnaire consists of two question forms The first form comprises five questions are asked at a monthly frequency. They refer to : Past development of the business situation Past and expected demand developments Past and expected employment developments The second form of fourteen questions is part of the quarterly questionnaire and is used in January, April, July and October. They refer to past and future assessments of : Operating income Operating expenses Profitability Capital expenditure Competitive position
Financial Services Survey There are three different answer categories: increase/improve, remained unchanged, and decrease/deteriorate. In particular, response (in the time ? survey) to the question which asks expectations regarding the demand for services (turnover) in the next three-month period is compared with response (in the time ? + 3 survey) to the question which asks about realizations in the past three-month. Therefore, answers to expectation questions (at time t) and realization questions (at time t+3) will cover the same period. This property allows us to analyze expectation errors. Development over the last 3 months (t+3) ??? Increased Remained unchanged Decreased 0 - 1 - 2 Increase Expectations 1 0 - 1 Remain unchanged over the next 3 months (t) 2 1 0 Decrease
Methodology Total uncertainty is the sum of weighted squared institution- specific errors: ? 2, where ??? is the weight of financial institution ?, ??= ?=1 ?????? and ? ?=1 ???= 1
Methodology Increase in the uncertainty measure ?? can stem from two different factors 2+ ??. (by identity) ??= ?? ? The aggregate effect, ??= ?=1 An increase (or a decrease) in the mean expectation results in higher ??. When an unpredictable aggregate shock hits the economy most participants will face an error in their forecasts and hence ?? will increase. The discrepancy. ??= ?=1 ?????? ?? When participants make different errors, this means that there are idiosyncratic shocks. Therefore, the discrepancy is defined by the weighted squared differences of expectation errors from the mean errors Total uncertainty is the sum of (squared) aggregate factors and idiosyncratic factors. ??????. ? 2.
Results Fed announcement in May 2013 Also starting late May, a wave of demonstrations against the government and civil unrest, which is called Gezi protests, occurred and lasted a couple of months in Turkey. Under these circumstances, capital inflows to Turkey have also weakened and uncertainty about what would happen in domestic financial markets increased significantly Our measure, ??, is also increased by eight times m-o-m and by almost 200 percent with respect to early 2013 averages.
Results Domestic political risk increased substantially due to corruption claims against the government in October and November 2013. ??increased by 40 percent Local elections in late March amid corruption claims and increased political uncertainty. ??reached its highest level in March 2014
Results In June 2014, war and terrorism fears in Ukraine and Iraq were the main driving force in increased uncertainty. Turkey has significant economic relations with these neighbors and, before June 2014, Iraq had become the biggest export market for Turkish goods. Increased violence weighed heavily on export figures and raised concerns about the future. Amid these tensions, ?_? has increased to its high levels again.
Results Discrepancy (idiosyncratic factors) is the main driving factor for total uncertainty (for Turkish financial sector in the given period of time). Somewhat surprising All events discussed are economy-wide Nevertheless, the biggest hike that happened in February 2014 can be explained by idiosyncratic factors as most polls diverged before the local elections and this could shape expectations of institutions differently.
Methodology One question: Are these hikes represent level shocks rather than uncertainty shocks? That is, level effects, should be taken into consideration as well. We compute balances: ? ??= ?=1 ??? : the response given by participant ? at time ? and can take values 1,0, 1, for will increase, will remain unchanged and will decrease. ?? is an indicator of, on average, how participants expect their future turnover. ??????,
Results Level effects move in line with uncertainty only for February 2014. Before the local elections, expectations deteriorated and this could be a complementary factor in the rise of ??. Other times, increases in total uncertainty are not accompanied with adverse level shocks. For example, expectations remained stable in May 2013 when Fed announcement on its asset-purchasing programme increased ?? significantly. Level shocks are important but they are not necessarily the main factor in ??developments
Conclusion 1. Uncertainty measure derived from forecast errors can be used as a good proxy for uncertainty Big hikes at times when most important recent events in Turkish and international financial sector. Robust when controlled for level effects 2. Discrepancy is the main driving factor for total uncertainty Only visual evidence due to lack of a long time series. A more proper empirical research is forthcoming
CENTRAL BANK OF THE REPUBLIC OF TURKEY aytac.erdogan@tcmb.gov.tr
