The Global Financial Crisis was a unique phenomenon that not only affected many millions of people and tens of thousands of companies at the time but also impacted the global financial system and the future of financial institutions and services they provide. In order to better understand the issues, and to avoid repeating the same situations in the future (or, at least, mitigate the consequences), it is always important to learn the root causes. In the following paragraphs, the most important aspects of the crisis will be reviewed from the risk perspective: what gaps in risk assessment caused the crisis, what has improved in risk-management approaches since then, and how shocks change the financial-service landscape.
Lessons from the 2008 Financial Crash
Main Causes
Undoubtedly, the reasons why the 2007-2009 Crisis began and was so severe are complex and lie in many different areas of finance, lifestyle, and psychology. However, in terms of risk management, we can briefly summarise the most important systemic and industry players’ faults to the following list.
Most of these issues were mentioned in the 2009 Senior Supervisors Group (SSG) report [1]:
- Regulators and credit rating agencies underestimated the risk of sophisticated derivative products;
- Significant delay in the government/regulation response;
- Poor liquidity and market risk management in large financial institutions;
- Failure of top management to understand and choose appropriate levels of risk;
- The conflict of interests between managers’ compensation policies and risk control;
- Risk managers were not independent;
- Weak tech infrastructure of risk identification;
- The high cost of qualified risk managers due to risk management being insufficiently developed as a profession.
The Problem of Credit Rating Agencies
Credit rating agencies are a crucial institution and a useful instrument to tackle the information asymmetry between lenders and borrowers. However, as these institutions are commercial organisations, there always is room for a conflict of interests. The “big three” agencies (S&P, Fitch, and Moody’s), which are used as a reference by most banks and hedge funds all around the world, are fairly criticized for working on the “issuer pays” model [10]: when an issuer of a financial product wants a rating, it pays to an agency, creating a moral hazard.
In 2007, thousands of structured debt products, such as CDO and MBS, got the AAA rating from “big three” agencies – the highest credit rating possible. So, when the structured products began to fall in price (because of mass borrower defaults), the agencies had to swiftly downgrade those unreasonably high products’ ratings. Considering the ratings given by these agencies would be largely influenced by the risk of default, the fact that supposed “AAA-rated” companies had such mass defaults illustrates clearly how the associated moral hazard of credit ratings given by corporate entities can create a misrepresentation of risk. This led to the issue that became one of the symbols of the Global Financial Crisis — the problem with illiquidity:
Liquidity risk issue. Since structured credit products were graded on the same scale as bonds, many banks used corporate bonds with similar ratings as a proxy in their credit-risk models [9]. However, once ratings of those very popular structured products began to fall substantially (sometimes from the highest grade to the default grade), this launched a domino effect:
- Funds that can hold only AAA-rated products had to sell downgraded ones;
- Because of margin calls, hedge funds had to get rid of leveraged positions, selling them to the illiquid market;
- Market players lost the understanding of where the fair prices were;
- Banks had to re-estimate credit risks, tighten up their credit limits, and raise their credit rates;
- Households could not refinance their debts because of high rates and therefore defaulted, adding fuel to the fire of credit derivatives and contributing to the vicious cycle of raised rates and defaults.
After the 2008 Global Crisis, rating agencies made lots of efforts to regain the trust of society and market players. However, the main conflicts of interest remain the same: non-transparent assessment methodologies and the nature of their business model [11].
Risk Modeling Issues
The 2008 Crisis was a real test of risk policies and models across the industry — it demonstrated that the approaches to risk needed a deep review because assumptions, that earlier seemed to be stable and logical, were broken in what felt like a single moment. Unsurprisingly, international rating agencies were not the only “bad guys” who turned out to be wrong (intentionally or not) with their risk-assessment models. Thousands of financial institutions had very similar issues. One of the important mistakes was ignorance of the interconnection phenomenon (the connection between products, markets, and between risk types):
Default correlation. During the crisis, more than 13000 AAA-rated tranches of structured products defaulted one after another [8]. Such a unified failure was not expected by market players, as they did not consider the crisis behaviour of correlation between different debt-related products with high credit ratings: the correlation between collateralized debt obligations, mortgage-backed securities, and other structured debt derivatives jumped dramatically. The ethical problem also was in the game because the use of low correlation in the companies’ risk models made it possible to take more risks and more instant profits: “Even if many players knew that correlation estimate to be a total fluke, they could not afford to yell the truth. The wrongful estimates were making them too much money.” (Pablo Triana, Financial Times, Correlations that you can trust[4]).
Correlation between markets. During the long history of financial shocks, there was a reappearance of the effect of intermarket correlation increase [7] — in crisis situations, different markets fall consequently, even if in a normal state they are not that much correlated. In the 2008 crash, this effect was the most noticeable compared to previous crises. Later studies even note that, after the 2008 turmoil, when some stock market correlations returned to the normal position, others still remained with higher co-movement levels than before [3].
Risk endogeneity problem. Not only products and markets can show a strong correlation, but the risks themselves are also tightly interconnected [2], and this effect was also not seriously considered. In fact, risks cannot be simply separated into Market, Liquidity, Credit, and Operational risks, like they are split in study books. As an example from what happened in 2008: the realization of credit risk triggered liquidity risk, which pushed market and operational risks. The impact of one type of risk on another should be considered in a company risk framework.
Another mistake was the too strong reliance on probability-based risk estimators [2]:
Standard Value-at-Risk measure problem. The Standard VaR approach was widely used as the main market risk measure by private and institutional investors. However, this approach was (and still is) highly criticized by Nassim Taleb [5] and other authors. The fair critique is based on the facts that firstly, standard VaR uses Gaussian distribution (which does not exist in the real market), and, secondly, standard VaR does not cover anything above the confidence interval: you can have a 99.9% probability of risk lower than 5% of your capital but just one case from the remaining 0.1% may cost 1000% of the capital. 2008 made clear that simple probability-based Value at Risk is not robust as the only measure of market risk. To be more informative, the VaR measure can be adjusted (Conditional VaR is an example [24]), and it should be used along with other market-risk measures (Expected Shortfall, EVT-measures, etc.) and stress testing.
Government Response
Looking at the timeline of the global-crisis [12], it is noticeable that governmental actions of providing extra liquidity to markets started only at the end of 2008 and early 2009 when the first serious alarms about the possible crash appeared in 2007. Nevertheless, the measure of providing extra liquidity, Quantitative Easing, turned out to be an effective instrument for crisis treatment [13].
Regulatory Aspect and Risk Practices’ Changes
After 2008, the financial-services landscape was significantly changed. Most countries improved their laws about investment, trading, and financial institutions. In the US, the most game-changing post-2008 law was the Dodd-Frank Act [14] — a regulation which covers a wide area of financial activities. The Act mandated the creation of additional entities for the control of financial stability and consumer protection, the creation of the Office of Credit Ratings (OCR) for overseeing credit agencies; and it also obligated banks to conduct special stress testing of potential impacts on their capital and introduced other risk-restrictive rules.
As to the international level, the Basel Committee (BIS) expanded their membership, increasing their power, and issued the Basel-III framework [16], which has become the new regulatory standard for international banks.
To investigate and tackle crisis-related issues, a special international entity, the Financial Stability Board (FSB), was established [17]. The board works on monitoring and addressing risks to financial stability. Among other things, the “systemic importance” rule was introduced, where institutions which are recognized as systemically important have to follow stricter rules and are put under the microscope of greater regulatory attention.
Risk management as a subject and a profession has gained dramatically more attention. In autumn 2008, the Global Association of Risk Professionals (GARP) registered a record number of early registrations for the certificate exam (a more than 40% increase) [18]. Now, risk management is an important part of any financial education curriculum, and every financial institution has at least one risk specialist, which was not the case before 2008. Also, model validation, as an area of risk control, has become more meaningful and important.
COVID-19 Crisis
How It Differs From the 2008 Crisis
The COVID-19 crisis, no doubt, has some similarities with the 2008 depression: the instant market reaction, or how both of them were relatively unexpected. However, it probably has even more differences. First and foremost, the nature of the latest crisis definitely is related to interconnectivity, but the origin is not financial. Considering this pandemic has been truly global – affecting people in every country in the world – it is characterised by the highest level of uncertainty measured by the Index of Global Economic Policy Uncertainty since its development in 1996 [19]. The leap in unemployment rate in 2020 was also substantially higher than in 2008.
Additionally, during the coronavirus period, new unique challenges have been faced and are a new reality for market participants:
The reality of Negative Prices. Owing to the drop in demand for oil due to the global lockdown, for the first time in history, oil futures prices went below zero. On April 20, 2020, right before the expiration, the price of the May WTI oil futures contract dropped more than 100%, to settle at $-37.63 per barrel [21], which was thought impossible in practice by most traders and financial institutions. Even more surprised were some exchanges: it was technically impossible for their infrastructure to process negative prices; their margin requirements calculations also considered a maximum price decrease for a long position as 100%. For example, the Moscow Exchange had to suspend trading in the WTI contract [22] which led to losses for many traders who were not able to close their positions (later, there was an attempt to sue the exchange [23], but it did not succeed). The negative price possibility pushed market players to implement it in the risk models, trading and back-office software.
Increased Importance of Operational Risks. Compared to 2008, where regulatory, market, and liquidity risks were the main battleground on which this crisis was fought, during the COVID-19 crisis the risk was mainly focused on operational risks. The radical transformation of the work environment – with the tendency of shifting of all services online for the sake of safety – made it crucial to develop new policies and monitoring-and-control mechanisms to manage the operational risk connected to work conditions. Mass remote work needs data-safety management and operational-error control to receive much higher levels of attention due to the potential for data security issues and lack of in-person employee oversight.
Why the Coronavirus Crisis Was Softer
The main feature that saved the world from a financially catastrophic outcome in 2020 was the fact that the industry was fundamentally more prepared and still had the memory of 2008 fresh in its mind. The changes in risk frameworks and regulation helped to prevent – or, at the very least, in some cases, mitigate – huge losses and chain reactions in the global banking industry.
Governments also learned lessons from the 2008 crisis and reacted much quicker; they also already had tried-and-tested instrumentation to apply. For example, in the US, the Federal Reserve had already initiated liquidity and funding operations by 15 March 2020 [20], barely four days after the coronavirus spread had been declared a global pandemic. The US Congress soon followed with direct help programs to support citizens...
The more stable and regulated financial system, multiplied by gigantic amounts of liquidity thrown to the markets to support them, made the financial aspect of the global pandemic not as bad as it could have been.
Conclusion
The main lesson of these two crises is that, in the era of globalization, everything is interconnected. Even though the mechanisms and laws of that interconnectivity are not yet adequately understood, they exist and should be considered by risk model designers and validators.
Nevertheless, the effect of shocks can be succinctly articulated as “What does not kill us makes us stronger” — history shows that, after every shock, the global system adapts and becomes more durable. With increased regulatory control, enhanced risk-management frameworks, and improved self-assessment, the financial services system becomes more stable.
At the same time, combining new technologies and the tendency towards globalization and interconnectivity in all spheres of life will, undoubtedly, create new triggers for financial crashes and new sources of risk.
References
1. Senior Supervisors Group, Risk Management Lessons from the Global Banking Crisis of 2008, https://www.fsb.org/2009/10/r_0910a/, October 2009
2. Charilaos Mertzanis, Risk Management Challenges after the Financial Crisis, https://www.researchgate.net/publication/259551142_Risk_Management_Challenges_after_the_Financial_Crisis, 2013
3. Yonghong Jiang and others, The Financial Crisis and Co-Movement of Global Stock Markets—A Case of Six Major Economies, https://www.mdpi.com/2071-1050/9/2/260, 2017
4. Pablo Triana, Correlations that you can trust, https://www.ft.com/content/6ca10d5a-9681-11df-9caa-00144feab49a
5. Hearing before the Subcommittee on Investigations and Oversight Committee on Science and Technology: The Risks of Financial Modeling: VaR and the Economic Meltdown, https://www.govinfo.gov/content/pkg/CHRG-111hhrg51925/pdf/CHRG-111hhrg51925.pdf
6. Pilar Abad, A comprehensive review of Value at Risk methodologies, https://www.elsevier.es/en-revista-the-spanish-review-financial-economics-332-articulo-a-comprehensive-review-value-at-S217312681300017X , 2014
7. Leonidas Sandoval Junior and others, Correlation of financial markets in times of crisis, https://arxiv.org/abs/1102.1339 , 2011
8. Jordan Nickerson, John M. Griffin, Debt correlations in the wake of the financial crisis: What are appropriate default correlations for structured products?, https://www.jgriffin.info/wp-content/uploads/2017/11/corr_pub.pdf, 2017
9. Michel Crouhy, Risk Management Failures During the Financial Crisis, October 2011
10. CFR, The Credit Rating Controversy, https://www.cfr.org/backgrounder/credit-rating-controversy, 2015
11. Truthout.org, The Indisputable Role of Credit Ratings Agencies in the 2008 Collapse, and Why Nothing Has Changed, https://truthout.org/articles/the-indisputable-role-of-credit-ratings-agencies-in-the-2008-collapse-and-why-nothing-has-changed/
12. Mauro F. Guillén, The Global Economic & Financial Crisis: A Timeline, https://ypfs.som.yale.edu/library/global-economic-financial-crisis-timeline
13. Kimberly Amadeo, Explaining Quantitative Easing (QE), https://www.thebalance.com/what-is-quantitative-easing-definition-and-explanation-3305881 , June 2021
14. Sean Ross, What Major Laws Were Created for the Financial Sector Following the 2008 Crisis?, https://www.investopedia.com/ask/answers/063015/what-are-major-laws-acts-regulating-financial-institutions-were-created-response-2008-financial.asp, 2020
15. Kelly Anne Smith, How The Dodd-Frank Act Protects Your Money, https://www.forbes.com/advisor/investing/dodd-frank-act/ , 2020
16. BIS, Basel III: international regulatory framework for banks https://www.bis.org/bcbs/basel3.htm
17. Financial Stability Board, Post-2008 financial crisis reforms, https://www.fsb.org/work-of-the-fsb/market-and-institutional-resilience/post-2008-financial-crisis-reforms/
18. The Global Association of Risk Professionals, GARP Announces Record-High 2008 FRM(R) Exam Registration, https://investegate.co.uk/the-global-association-of-risk/prn/garp-announces-record-high-2008-frm-r--exam-reg---/20080501184500NO108/ , 2008
19. Atlantic Council, Can we compare the COVID-19 and 2008 crises? , https://www.atlanticcouncil.org/blogs/new-atlanticist/can-we-compare-the-covid-19-and-2008-crises/
20. Richard H. Clarida and others, The COVID-19 Crisis and the Federal Reserve’s Policy Response, https://www.federalreserve.gov/econres/feds/the-covid-19-crisis-and-the-federal-reserves-policy-response.htm
21. Commodity Futures Trading Commission, Interim Staff Report: Trading in NYMEX WTI Crude Oil Futures Contract Leading up to, on, and around April 20, 2020, https://www.cftc.gov/PressRoom/PressReleases/8315-20
22. Moscow Exchange announces changes to trading in the Light Sweet Crude Oil futures contract nearest to expiry, https://www.moex.com/n28144/?nt=201
23. Investors sue Moscow Exchange over losses on oil futures trading halt, https://www.nasdaq.com/articles/investors-sue-moscow-exchange-over-losses-on-oil-futures-trading-halt-2020-06-05
24. Adrian, Tobias, and Markus K. Brunnermeier, CoVaR, https://scholar.princeton.edu/sites/default/files/markus/files/covar_0.pdf, July 2016
© Ivan Samkov, 2021