• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Research in Financial Economics: High-frequency trading, optimal trading strategies, idiosyncratic uncertainty in asset pricing and financial stability of banks

Priority areas of development: economics
2017
The project has been carried out as part of the HSE Program of Fundamental Studies.

Goal of research

1.«Debt, Recovery Rates and the Greek Dilemma»  We show that when additional funds released to the debtor (Greece), via debt restructuring, are used efficiently in pursuit of a practicable business plan, then both debtor and creditor can benefit.

2.«Firm-specific uncertainty around earnings announcements and the cross-section of stock returns» We examine whether the firm-specific component of ex-ante earnings-related uncertainty is a priced characteristic for stocks.

3.«High-Yield Insured Deposits and Financial Stability» We study the risks and consequences of aggressive deposit pricing in the deposit market on banks’ financial intermediation model, profitability, and probability of failure.

4.«High Frequency Trading and Market Quality» We construct a model in order to analyze an impact of the high frequency trading (HFT) on various characteristics of market quality, such as liquidity, trading cost, and price efficiency.

Methodology:

1. «Debt, Recovery Rates and the Greek Dilemma»  We examine a dynamic two country model calibrated to Greek and German economies and support two-steady states, one with endogenous default and one without, depending on creditors’ expectations. In the default steady state, debt forgiveness lowers the volatility of both German and Greek consumption whereas demanding higher recovery rates has the opposite effect. In a second order approximation of the model, conditional welfare analysis shows that a policy of immediate leniency followed by harsher terms as the economy grows is beneficial to both creditors and debtors.

2. «Firm-specific uncertainty around earnings announcements and the cross-section of stock returns» We use option-implied expected earnings announcement day jump variance as the measure for uncertainty and extract the firm-specific component by isolating the portion, correlated with the squared market risk sensitivity. Time series asset pricing tests, applied to the portfolios sorted on firm-specific earnings-related uncertainty, yield that this measure has explanatory power for stock returns: We find economically and statistically significant returns in excess of both the Carhart (1997) model and Fama-French (2015) model predictions for a portfolio which loads on firm-specific earnings-related uncertainty. We show that this result is driven by large cap stocks and is independent of operating profitability.

3. «High-Yield Insured Deposits and Financial Stability» The methodology of the first part of the paper is a multidimensional panel regression with fixed time and banking effects. For the second part of the paper, we use the methodology of logistic regression, where the discrete dependent variable takes value one if the banking license was withdrawn by the regulator and zero otherwise.

4. «High Frequency Trading and Market Quality» We propose a new method to deriving the dynamic trading strategies, different from the dynamic programming approach employed in Back, Cao and Willard (2000) model. Technically, our method leads to the close set of deterministic integro-differential equations describing the trading strategies of informed and market makers, whereas the dynamic programming method requires to solve the system of Hamilton-Jacobi-Bellman partial differential equations, which is much less tractable analytically or numerically. Making use of our new method, we were able to give a much simpler derivation of the result of Back, Cao and Willard (2000). Next, we extend their model to the case of finite number of oligopolistic profit-maximizing market makers, and derive a description of optimal strategies in terms of a close system of deterministic integro-differential equations.

Empirical base of research:

1. «Debt, Recovery Rates and the Greek Dilemma» Model is calibrated on macroeconomic data for Greece and Germany. We have calibrated our economy according to the values provided in previous studies at a quarterly frequency. First we discuss the calibration with respect to the Acyclical recovery rate policy. The secured interest rate is 2% per annum while the unsecured rate is 12.9% per annum. This compares with the average Germany 10 year yield from January 2010 to October 2014 of 1.916% and Greek 10 year bond yield from January 2010 to January 2015 of 12.81%. The recovery rate on Greek debt was taken from Vrugt (2011) who estimated that the recovery rate on Greek debt to be between 40 and 60 - we have taken the midpoint of 50. This is also the number documented in the cross-country findings of recovery rates found in Benjamin and Wright (2013). The volatility of default rates was taken from the Standard and Poors Ratings Direct study “Default, Transition, and Recovery: 2012 Annual Global Corporate Default Study And Rating Transitions”, which gave the mean and standard deviation of annual European Corporate Speculative-Grade Default Rates as 3.1% and 3.46% respectively.

2. «Firm-specific uncertainty around earnings announcements and the cross-section of stock returns» We obtain options data from IvyDB US Optionmetrics. Fixed time-to-maturity implied volatilities are calculated by Optionmetrics as interpolation of implied volatilities of closest options with respect to time-to-maturity and moneyness. Stock price data, market capitalization and earnings announcement dates are coming from Datastream. Returns on the market, size, value, momentum, operating profitability and investment intensity portfolios are taken from the online data library of Kenneth French. We use all stocks, which were constituents of the Standard & Poors Super Composite 1500 index on January 31, 1996, June 30, 2005 and September 30, 2015. After dropping stocks which do not have corresponding options data or earnings announcement dates in 1996-2013 we are left with 1740 stocks for the analysis.

3. «High-Yield Insured Deposits and Financial Stability» The final sample consists of an unbalanced panel of 555 unique banks that we trace from the end of 2012 to June 1, 2017. Our financial data come from the banks’ balance sheet and income statement disclosures to the central bank and cover eleven quarters, from Q4 2012 to Q2 2015.  Bank failure data come from official central bank releases on bank closures that we trace up to the June 1, 2017.  We used this institutional feature to extract bank-level deposit product terms through financial portal of Russia’s premier Internet search engine www.yandex.ru. 

4. «High Frequency Trading and Market Quality» in this part of our research project the theoretical model is built and its properties are studied.

Results of research:

1. «Debt, Recovery Rates and the Greek Dilemma» We showed that a policy of immediate leniency followed by harsher terms as the economy grows is beneficial to both creditors and debtors.

2. «Firm-specific uncertainty around earnings announcements and the cross-section of stock returns» We examine whether the firm-specific component of ex-ante earnings-related uncertainty is a priced characteristic for stocks. We use option-implied expected earnings announcement day jump variance as the measure for uncertainty and extract the firm-specific component by isolating the portion, correlated with the squared market risk sensitivity. Time series asset pricing tests, applied to the portfolios sorted on firm-specific earnings-related uncertainty, yield that this measure has explanatory power for stock returns: We find economically and statistically significant returns in excess of both the Carhart (1997) model and Fama-French (2015) model predictions for a portfolio which loads on firm-specific earnings-related uncertainty. We show that this result is driven by large cap stocks and is independent of operating profitability.

3. «High-Yield Insured Deposits and Financial Stability» Our results detect a number of fundamental risks associated with aggressive deposit-taking bank behavior. First, we find that high yield deposit-taking is associated with fast deposit growth and gain in the deposit market share. We also show that aggressive deposit taking policy is associated with higher loan growth and investment into private securities and elevation of credit risk. This leads to shrinkage of the net interest margin. Finally, we find that banks which offer high-yield deposit products have higher likelihood of failure and draw down higher payments from the deposit insurance fund in case of failure. Collectively, our results reveal substantial funding and asset-side risks associated with high-yield deposit products in a market with a flat deposit insurance premium, aggressive competition for limited funds, and insured depositors’ disincentives to monitor banks’ behavior.

4. «High Frequency Trading and Market Quality» As a result of our project, we suggest a new solution method for these type of problems. Namely, we develop a new type of equilibrium for dynamic Kyle model using BSE approach.

Publications:


Goodhart C., Peiris U., Tsomocos D. Debt, Recovery Rates and the Greek Dilemma // Journal of Financial Stability. 2018. No. 26. P. 265-278. doi
Brown M., De Haas R., Sokolov V. Regional Inflation, Banking Integration and Dollarization // Review of Finance. 2018. Vol. 22. No. 6. P. 2073-2108. doi
Sokolov V., Solanko L. POLITICAL INFLUENCE, FIRM PERFORMANCE AND SURVIVAL / NRU HSE. Series WP BRP 60/FE/2017 "SERIES: FINANCIAL ECONOMICS". 2017.
Sprenger C., Lazareva O. CORPORATE GOVERNANCE AND INVESTMENT: EVIDENCE FROM RUSSIAN UNLISTED FIRMS / NRU HSE. Series WP BRP 160/EC/2017 "BASIC RESEARCH PROGRAM WORKING PAPERS". 2017.