RBI: Measuring Financial Stress in India – A study

Reserve Bank of India in its RBI Working Paper Series No. 09 dated September 22, 2020, published a paper on its website which measures the financial stress in India focusing on the construction of financial stress indices for the Indian financial system using market-based indicators and three different aggregation methods. The paper finds that these indices provide useful information, which is not captured by the market indicators, on the stress build-up in the financial system.

Let us understand the word financial stress which is often used in connection with banks. However, the authors define it as episodes when economic agents face extreme uncertainties and varying expectations of losses in financial markets. To measure financial stress, financial market information or balance sheet data is used. We agree when we are informed that various segments of financial markets contribute observable indicators to be measured as indices using appropriate statistical techniques. By calculating and studying financial stress, we may avoid systemic risk.

The world over, central banks measure financial stress by constructing the financial stress index (FSI) to monitor the functioning and resilience of the financial system. Such an index provides an aggregate measure of financial stability to the policymakers. FSIs may help in gauging systemic risks and the probability of occurrence of stress events.

By counter arguing, if we have the FSI, we may get early warning signals about impending financial stress, and also use them to actually study the functioning of capital markets.

Let us look at the study from the following website and learn more.

https://rbidocs.rbi.org.in/rdocs/Publications/PDFs/09WPS220920202604968D5E644ED59A394128EC9F5FBE.PDF

(“Measuring Financial Stress in India”) by Manjusha Senapati and Rajesh Kavediya, Economists from Reserve Bank of India.

The study consists of 25 pages with the following chapters:

• Introduction
• Literature
• Selection of financial markets and variables
• Construction of FSI
• Financial Stress Index for the Indian Financial System
• Financial Stress Index and the real economic activity linkages
• Dating the financial crisis episodes
• Conclusion

Being a mathematical and economic study with the latest techniques which most of us would not understand, I intend to narrate those concepts with the mentioning of the names of the methods used as and when required.

It starts with the question of financial stress as viewed from history. Some of the narrations on financial stress is worth mentioning:

1. The global financial crisis of 2008-2009 kindled the interest to study the stress in the financial system. Research started to focus on defining it and linking it with stress episodes.
2. It is a result of the interaction between vulnerable markets and endogenous shocks as per a study in 2010.
3. Balakrishnan in 2009 said that Stress was characterized by certain features such as exchange rate pressures or depletion of forex reserves and tighter bank lending.
4. Some more revelations: Some studies focused on a specific segment of the financial market – banking sector – and found that banking distress led to longer periods of economic distress as compared to that caused by distress in securities or forex markets (Cardarelli et al., 2011, Hanschel and Monnin, 2005).
5. However, in this paper, financial stress is defined, following Hollo et al. (2012) and Huotari (2015), as the stress that has already materialized and could result in widespread financial instability.

Let us learn what happened in the past.

Large shifts in asset prices, a sudden increase in risks, liquidity stress, and banking system stress were associated with increased financial stress. Past stress events were characterized by one or more of the following four elements, for example, (i) rising uncertainty about the fundamental value of assets, (ii) higher uncertainty about the behavior of other investors, (iii) increased asymmetry of information, and (iv) decreased willingness to hold risky and/or illiquid assets.

several FSIs for different countries or a group of countries have been used to describe system-wide stress in financial markets. Illing and Liu (2006) constructed a financial stress indicator for the Canadian economy. They constructed an FSI as a continuous measure of financial stress and covered equity markets, bond markets, foreign exchange markets, and the banking sector. Their FSI provides an ordinal and contemporaneous measure, capable of tracking stress events.

About India

In this paper, three FSIs are developed for the Indian financial system using three different aggregation methods, covering data on five different segments of Indian financial markets. The objective is to construct FSIs which can be useful in identifying 3 stress build-ups in the financial system. This paper adds to the scant literature available for the Indian financial market in the area. Additionally, it explores the relationship between FSIs and economic activity.

For India, Shankar (2014) constructed a Financial Conditions Index (FCI) using monthly data from January 2004 to August 2013. He uses data from four financial markets – money, bond, foreign exchange, and stock markets – to construct the FCI using the principal components analysis method. The index is found to have a high correlation with IIP and Gross Domestic Product (GDP). He argues that it would be better to take into account the financial conditions of all markets together instead of dealing with them separately to better deal with the information asymmetry.

Guru (2016) developed a composite Financial Sector Stress Index (FSSI) for the Indian financial system taking into account the currency market, the banking sector, and the stock market for the period April 2001 to December 2012. He claims that FSSI is a useful early warning indicator. Extreme stress events are identified using Extreme Value Theory (EVT).

 Gayen (2016) constructed FSI for India using the concept of multivariate distance and semivariance. The data on ten variables is grouped into three financial sub-markets that is the banking sector, the foreign exchange market, and the debt market. He argues that the proposed FSI is able to capture the financial market stress well.

Each episode of financial stress is unique originating in one or the other segment of financial markets. In this paper, the selection of financial markets is done so as to cover the whole financial system. Another important question relates to the number of indicators to be chosen from each of the financial markets to depict stress.

The selection of the final stage variables in this paper was made on the basis of their correlation with the real economy as proxied by the growth in IIP. The data period extends from January 2002 to September 2019. Massive coverage of 17 years of studies, revealing the secrets of financial stress over nearly three decades. An applaudable goal indeed!

Let us gravitate towards serious items chosen for the study.

• Money market-related sub-index components (from page 9 of the study) (1)
• Realized volatility3 of 3-month-interbank rate
• TED Spread (Interbank Spread)
• Debt market-related sub-index components (2)
• Realized volatility of 10-year government bond yields
• 10-year government bond yields spread over global yields
• Equity market-related sub-index components (3)
• Banking related sub-index components (4)
• Realized volatility of the banking sector equity index
• Banking sector beta
• Foreign exchange market-related sub-index components (5)
• Realized volatility of the exchange rate

Numbers 1 to 5 indicate the factors taken for study.

Construction of FSI

For the construction of FSI, each individual indicator selected from five financial markets was standardized and transformed to ensure comparability across the markets. Several studies have transformed the variables using their cumulative distribution functions (CDFs). The following heads were studied under the construction of FSI.

1. Variance equal weighting
2. Principal Components Analysis
3. Portfolio theoretic aggregation

Financial Stress Index for the Indian Financial System

The following charts were created to study FSI for the Indian financial system.

1. Chart 1: Money market FSI
2. Chart 2: Debt market FSI
3. Chart 3: Equity market FSI
4. Chart 4: Banking sector FSI
5. Chart 5: Forex market FS
6. Chart 6: Financial Stress Index

The revealing truths based on the study may enlighten us in a simple way: (I am constrained to quote from the report many times directly since the study based on the highest level of economic/statistical/mathematical models with complex formulae are beyond my comprehension and I just understand simple conclusions drawn by the experts by using the latest technologies and computer models)

• The money market sub-index (Chart 1) combines the realized volatility of the 3- month MIBOR and TED spread. Individual series in each sub-market is standardized and then the market-specific sub-indices are constructed using the averages for all constituent series in each submarket. The movement of money market FSI shows the highest level of stress in October 2008. This peak was higher than that reached in March/July 2007 due to extreme market volatility. Risk aversion by the investors and worries over credit stress resulted in TED spreads rising dramatically.
• The debt market sub-index (Chart 2) combines realized volatility of the 10-year Indian government bond yields and 10-year Indian government bond yields’ spread relative to the 10-year US bond yields. It is evident that in the year 2008-09, following the Lehman crisis, bond markets were highly stressed. The next peak was registered in September 2013 around the time of the taper tantrum.
• The equity market index consists of volatility and the Cmax of equity market index. The equity market FSI reached its peak in October 2008 at the time of the global financial crisis (Chart 3). As noted by Shankar (2014), the stress in the equity market reverted to its mean level irregularly.
• The banking sector’s FSI combines realized volatility and Cmax for the Nifty bank index and the banking sector beta. The peak stress was experienced at the time of the global financial crisis and at the time of the taper tantrum (Chart 4).
• The realized volatility and the Cmax of the exchange rate were combined to construct the forex market FSI. Stress in the forex market was high at the time of the global financial crisis in 2008 due to high volatility (Chart 5). Reflecting on the impact of the taper tantrum, the index reached its peak after May 2013 due to a sharp depreciation of domestic currency amidst high volatility (Shankar 2014). It was, however, short-lived due to the measures taken by the Reserve Bank of India and the Indian government.

What do we understand from the above conclusions from a layman point of view?

1. Highest level of stress in October 2008
2. Bond markets followed the collapse of the Layman crisis in 2008-2009.
3. The equity market followed the stress in October 2008, at the time of global crisis.
4. The global crisis again played a role in banking stress in 2008.
5. Sharp variations in the foreign exchange rate in 2013 were met head-on by RBI and brought stability in the market.

Now the coveted discussion on financial stress index as a whole

It is evident that the highs in the FSIs could be linked with well-known stress events.

All the three FSIs (compiled using different methodologies) peak during the crisis periods of October 2008 relating to the global financial crisis and August 2013 relating to the taper tantrum. In the initial years of 2002-06, moderate US-India 10- year bond spreads and relatively accommodative foreign exchange market conditions and the stable banking sector kept FSI low (except for around May 2004, due to uncertainties around the general election’s results and around March 2006 due to stock market crash following heavy selling by FIIs, retail investors and weaknesses in global markets)

 FSI witnessed its peak around September/October 2008 reflecting the underlying building up of financial stress around the global financial crisis marked by stress in all financial markets that were considered.

 FSI went up again in December 2011 because of exchange rate pressures. During June and August 2013, following the taper tantrum event, there were exchange rate pressures on account of rupee depreciation due to high net capital outflows. The US Fed’s indication that it may taper off the quantitative easing aggravated stress which got manifested in high FSI values in mid-2013.

Conclusion

The paper states its attempt as a foray towards the construction of financial stress indicators for the Indian financial system.

 It used 11 market-based indicators covering major segments of the Indian financial markets

 – money market, debt market, equity market, forex market, and the banking sector.

Three different methodologies – variance equal weighting, principal component analysis, and portfolio aggregation method were applied to construct the FSIs.

All three FSIs were found to be negatively correlated with real economic activity (IIP growth) and can be used as a leading indicator for predicting real economic activity.

 Also, the regression analysis suggested that the FSIs could be helpful in predicting IIP growth.

Periods in which FSIs are above the 90th percentile also witnessed a substantial decline in IIP growth.

 While identifying past financial stress, it was seen that systemic stress was present in the Indian financial system during the third quarter of 2008-09.

Yes, I am proud to introduce you to a highly researched paper on financial stress in the Indian economy over three decades, a pioneering effort by brilliant economists, namely, Manjusha Senapati and Rajesh Kavediya who have shown that RBI with its best human resource can unearth any economic turbulence and India can undoubtedly rely on its prowess for stepping into a new horizon. It is proudly proclaimed that the views expressed in the study belong to its researchers and not that of RBI, an indication of the brilliance and confidence of the economists.

*****

Disclaimer: All the information collected by me is my way of understanding the financial stress from the RBI website and obviously, neither taxguru.in nor RBI is responsible for my views. Please visit the web site quoted for understanding the complex research project.