"Information Processing
- Steve Hsu
- Professor of physics at the University of Oregon. Homepage. Archive (list of posts, by date and category).
Tuesday, June 09, 2009
Plight of the risk managers
I'd like to recommend this podcast interview with Riccardo Rebonato. Rebonato, the author of the prescient book Plight of the Fortune Tellers, written before the financial crisis, is a former physicist turned quant.
Rebonato does not mince words, pointing out the weaknesses of mathematical models, and noting that most quants, although mathematically sophisticated, often lacked deep knowledge about markets and banking (I assume he does not include himself in this group). In my experience many quants never questioned the basic efficient market assumptions underlying their models, although some certainly did -- in particular, those with trading experience.
Rebonato is polite, even urbane, but disagrees with Econtalk interviewer Russ Roberts on many important issues. The most important question, which Rebonato addresses immediately, is whether enlightened, self-interested managers of financial institutions can be relied on to properly manage risk. Regulators accepted, on faith, the self-regulating abilities and properties of a system managed by such people. Thus, one of the main ingredients in the crisis was the ideological (as opposed to political or financial) capture of regulators by efficient market proponents.
Other interesting topics covered are the divergent risk tolerances and interests of bond holders vs equity holders vs regulators of banks [1], and whether moral hazard (anticipation of a bailout) played a role in the crisis -- Russ, the anti-government libertarian, says yes. Rebonato says no, the story only makes sense if told at the institutional level, whereas individual incentives were different. I think Rebonato's logic is impeccable. It's more persuasive to me that incentive schemes which allowed huge compensation based on short term (ultimately illusory) gains were much more of a factor. (See Clawbacks, fake alpha and tail risk.)
[1] Footnote: see my earlier post on the vacuous Modigliani-Miller theorem. I recently learned from Vernon Smith's memoir (see pages 230-231 and 276) that he has similar opinions. Google books link; also search under "MM".
Rebonato does not mince words, pointing out the weaknesses of mathematical models, and noting that most quants, although mathematically sophisticated, often lacked deep knowledge about markets and banking (I assume he does not include himself in this group). In my experience many quants never questioned the basic efficient market assumptions underlying their models, although some certainly did -- in particular, those with trading experience.
Rebonato is polite, even urbane, but disagrees with Econtalk interviewer Russ Roberts on many important issues. The most important question, which Rebonato addresses immediately, is whether enlightened, self-interested managers of financial institutions can be relied on to properly manage risk. Regulators accepted, on faith, the self-regulating abilities and properties of a system managed by such people. Thus, one of the main ingredients in the crisis was the ideological (as opposed to political or financial) capture of regulators by efficient market proponents.
Other interesting topics covered are the divergent risk tolerances and interests of bond holders vs equity holders vs regulators of banks [1], and whether moral hazard (anticipation of a bailout) played a role in the crisis -- Russ, the anti-government libertarian, says yes. Rebonato says no, the story only makes sense if told at the institutional level, whereas individual incentives were different. I think Rebonato's logic is impeccable. It's more persuasive to me that incentive schemes which allowed huge compensation based on short term (ultimately illusory) gains were much more of a factor. (See Clawbacks, fake alpha and tail risk.)
Dr. Riccardo Rebonato
Riccardo is Global Head of Market Risk and Global Head of Quantitative Research and Quantitative Analysis for Royal Bank of Scotland based in London. Prior to joining the Royal Bank of Scotland, he was Head of Complex Derivatives Trading Europe desk and Head of Derivatives Research at Barclays Capital, where he worked for nine years.
Riccardo is a Visiting Lecturer at Oxford University in Mathematical Finance and Adjunct Professor at the Tanaka Business School, Imperial College, London.
Before joining the financial world, Riccardo was a Research Fellow in Physics at Oxford University (Corpus Christi College) and, before that, Visiting Scientist at Brookhaven National Laboratory.
Riccardo is the author of the books Plight of the Fortune Tellers ('07), The Perfect Hedger and the Fox (Wiley ’04), Modern Pricing of Interest-Rate Derivatives (Princeton University Press ’02), Interest-Rate Option Models (Wiley ’96,’98), Volatility and Correlation in Option Pricing (Wiley ’99). He has published several papers on finance (option modelling, computational techniques, risk management) in academic journals. He is a regular speaker at conferences worldwide.
[1] Footnote: see my earlier post on the vacuous Modigliani-Miller theorem. I recently learned from Vernon Smith's memoir (see pages 230-231 and 276) that he has similar opinions. Google books link; also search under "MM".
And from EconLog:
Riccardo Rebonato of the Royal Bank of Scotland and author of Plight of the Fortune Tellers talks with EconTalk host Russ Roberts about the challenges of measuring risk and making decisions and creating regulation in the face of risk and uncertainty. Rebonato's book, written before the crisis, argues that risk managers often overestimate the reliability of the measures they use to assess risk. In this conversation, Rebonato applies these ideas to the crisis and to the challenges of designing effective regulation.
Time: 1:02:22
Size: 28.6 MB
Right-click or Option-click, and select "Save Link/Target As MP3.
Readings and Links related to this podcast
| Podcast Readings HIDE READINGS |
|---|
About this week's guest: About ideas and people mentioned in this podcast:
|
Highlights
| Time | Podcast Highlights HIDE HIGHLIGHTS |
|---|---|
| 0:36 | Intro. [Recording date: May 22, 2009.] Book, risk and uncertain, decisions in an uncertain world; written beforehand but prescient about the economic crisis. History of finance and housing in the United States, how banking changed. In 1990s, up to end of 2006; hiccup of 1998, after the fact purely contained to financial sector; within 6-9 months a blip. Regulators who look after soundness of banking system came to rely on quantitative models and the banking system being the forefront by making use of these quantitative models. Two steps back from technicalities, philosophy of faith in self-regulating properties of the system, managed by enlightened self-interested mangers and CEOs; skilled. Quants came from background in mathematical finance, not deep thinkers, not deep knowledge of the banking system. Supposed to guarantee that sufficient capital would be set aside to act as safeguard against unlikely events. We know now that that didn't happen. Book: we are not in a position of saying anything about extremely rare events. Might collect daily, monthly, quarterly data, spanning a few years--information about the last few years. Even with data going back to the 1950s or 1940s, leap of faith to say they are relevant to present market conditions. What about events every 100, 1000, or 4000 years? A 99.975 percentile one year horizon corresponds to one event that should not happen more frequently than once every 4000 years. Quants might have sophisticated theory, but how do we know the data are sufficiently homogeneous--belong to the same patch of history--to be giving information relevant today as opposed to 3 years ago. Similar point made on EconTalk, talking about the Great Depression: no consensus about how we got into it or how we got out of it because it's a one-time event. Easy to have a theory, hard to be confident that it applies today. Humility rather than hubris. |
| 7:59 | Little more technical: one technique that came to be popular in the banking is Value at Risk. What is idea behind it and why did it become popular? Measure of risk with some nice features: expressed in units of money. If your value at risk one day, 95th percentile, is $20 million, it means you should exceed a loss of $20 million only 5 days out of 100, so basically one day in 20. Won't be a perfect measure, but gives idea of day-to-day volatility when nothing extraordinary happens. For a trader or for a bank as a whole. Look carefully at the time-frames we are talking about. Can I infer the type of losses I might typically incur over that period, one month or a few months? Yes. Supposed to be used to determine capital; but capital should be there to absorb losses that occur not just once every 20 days, but once every hundreds of years; any time to protect the solvency of the institution. Before the regulators slapped on a multiplier; didn't ask for more remote percentiles; just take a safety factor. Safety factors have illustrious pedigree. Physics, engineering: use same formula calculate the thickness of steel ropes (cables) in elevators; but at the end of the day, they multiply by 10. Calculations for landing gear of aircraft multiply by 1.1; always a safety factor. Sensible because it allowed the bank to calculate a statistical number that was meaningful; but then a multiplier. Goal of regulator: to prevent insolvency or disaster because if it spreads the entire financial system could be at risk, systemic problems, runs on the bank, recent problems. What nest egg, what buffer do you have to keep so that in the worst case scenario, your firm still survives? |
| 13:40 | Complicated question: inherent tension between bondholders and stockholders in a bank, and between regulators and owners. How big would the buffer be without regulation? How much economic capital--the buffer--should a bank keep to entice its investors, both equity and bond, to be comfortable? Not precise; what are some of the conflicts? Lot of thinking after the crisis about the alignment of interests even of any enlightened owner and the interests of society. Economic capital: the nest egg, but how severe an event? Folklore came up with a number: the 99.975 percentile--events that should happen not more frequently than once every 4000 years. Really bad luck, recent crisis! How did that number come up? Example: I am a bank with a certain rating. How many AA-rated institutions went from double-A to default in one year? Probably a handful. Divide that over total number of companies, get approximately one in 4000. So to convince bondholder that I am a AA company I should set aside enough capital in order to convince them I will only go bust once every 4000 years. Align capital I am holding with observed frequency of default of a AA company within one year. Signaling device to tell regulators: I am double-A. Regulators shouldn't be asking for a great spread, because I am AA. In case of a bondholder, the only thing of interest is solvency--just want their money back. Why not go for triple-A? Because the shareholders care about the return on capital? So to convince bondholder that I am AAA, I'd have to hold so much capital that the shareholder would not get a great return for each dollar of that capital. Compromise. For most banks, the equilibrium is around AA level. Bondholder, short a put, only loses money if the bank defaults; but if things go extremely well, he doesn't share in the upside. Shareholder likes volatility, long a call. Recent crisis has made very clear that a shareholder can be perfectly rational but willing to accept a level of risk that a regulator may not like. June or July 2007, shareholder of one of the banks that eventually got in trouble looks at his portfolio: would have to need a repeat of the 1930s to take a big loss; to insure himself against that possibility he'd have to lock in losses that most likely he won't have to bear. Perfectly acceptable to take that risk. Regulators' view: in that calculus you haven't taken into account the systemic risk. Regulators' view is to step in to not allow taking on that level of risk. Those investors had extremely high rates of return during 2003-2006. If someone had said this is getting a bit out of hand, your competitors wouldn't have been doing the same. In 2006, plenty of signs, liquidity was too abundant, risk premia compressed too much, asset prices too high. Fracture line of an earthquake; people knew it had to give, not sustainable for a long period of time; but where and when the crack would occur. Stepping out too early is disastrous. Greenspan, irrational exuberance, uttered around 1994, 1996. Pulling out would mean having to explain to the shareholders what you are doing when everybody else is becoming rich. |
| 23:35 | Competitive market issue, also a social issue. Small world, investment banks; events and parties where your colleague is doing very well, you look the fool in the short run if you pull out. Central question, heard that story before: that's the music that's playing, so if you are there to dance, you have to dance to the music. Everyone's investing in these mortgage-backed securities, etc., but it's the only game in town. Yet, there were people who did not play along. Range across institutions of how much exposure. Any speculative thoughts on why institutions differed? Bear Sterns vs. J.P. Morgan. Some institutions big in this area for a long period of time. Late-comers fared less well. Goldman Sachs, JP Morgan. Different levels of inventory reflected different stages of getting into this business. Could have been emerging markets, which were extremely resilient. Summer 2007, wobble in emerging markets; head trader described it as a week of panic buying of emerging market paper! For same amount of risk, less and less compensation being paid. Some have said this wouldn't have happened had the investment houses never gone public. Asymmetry of incentives facing traders, executives on one hand and investors on the other; not playing with their own money. From point of view of a private firm, it would have been a reasonable risk to take. Regulators different, system; level playing field. If I step off the bandwagon, disadvantage to a more reckless person or institution who remains on; people will have withdrawn their funds from me. Asset fund manager in the United Kingdom stepped out of the dot-com bubble a bit too early and got fired three weeks before the bubble burst. Difficult to swim against the tide; regulator should intervene to level the playing field, you will all be inhibited from taking this kind of risk. |
| 30:15 | Two questions. One: amount of leverage in play here is one reason these gambles were so large. One thing to say you had a bad quarter or bad year; but why did you take such a roll of the dice that these losses in the U.S. housing market destroyed your legendary company? Simple answer: they were so leveraged--they had borrowed so much to finance that gamble that losing the gamble didn't just mean a bad quarter, but death. Point number two: People who swim against the tide, standard argument is investor psychology is such that no one really cares about risk. But in current environment, psychology might thus go the other way. Akerlof and Shiller book, Animal Spirits, talks about the importance of sentiment; reversals of sentiment affect markets and real economy. How do we get out of the current environment--v or double-v as shape of the downturn? Very uncertain, unstable state. A week ago, stock prices were going up, people saying we'd get out very quickly; a week later, a bit of negative information. Might become more cautious if protracted period of decline. If plan A works and we get out with a v-shaped recovery, expect risk aversion will disappear much more quickly than we think possible today. Good that we get out quickly, but same behavior will result again in the same risks and problems. Where will that liquidity come from? |
| 34:51 | Basel II, regulatory arbitrage. Basel II was trying to make the capital a bank has to set aside more in line with the risk. Basel I, formulaic, simple rules for capital that did not reflect the risk taken by banks. Credit-worthiness of the bonds you hold--government of OECD countries considered very, very safe; banks of those countries considered very safe; little differentiation between lending to a double-A company and lending to a corner shop. Set aside $1000 to either the latter has higher return for apparently same risk. Basel II was attempt to remedy that. Overconfidence in statistical techniques' abilities to quantify risk. How did the ratings agencies come into this regulatory arbitrage issue? Set-up was role of the ratings agencies with securitization--process whereby banks accumulate assets, e.g., mortgages, package them, and by virtue of the diversification create a portfolio less risky than any of the individual items. Tranche the portfolio in different sections: first section bears risk of first losses, so gets more return if things go well; all the way up AAA. Ratings agencies were to check amount of diversification to be sure it was up to the right level. Human nature, and conflict of interest--rating agencies only got paid if the securitization went through. Agencies worked with the investment houses to meet the criteria for AA, AAA, and A ratings. Teacher sets test but goes over questions with the students the day before. Portfolios met the specific requirements, but not the best way to set up system. But someone who goes to hire a student from that kind of school knows to be skeptical. Everyone knew this about the ratings agencies. Value of franchise: only 18-24 months ago, enlightened self-interest of market players was viewed as best safeguard for financial system. If the ratings agency debase the currency of their name, they will lose credibility by giving away AAA for nothing. Self-regulating efficient market. |
| 41:10 | Mark to market: what role did it play in the crisis? Whole problem or irrelevant? Controversial topic. Not fair to point to that. Like saying if I didn't have a thermometer this patient wouldn't have a fever. Dynamics of mark-to-market forcing certain actions. Nexus where it got things into trouble was that when securities got downgraded, certain institutions had to sell them. This caused forced unwinding, Special Interest Vehicles (SIVs), nobody talks about it today. Mark-to-market is good where there is a willing buyer and a willing seller. Equilibrium is good information about the fundamental value of what is bought and sold. If unbalanced selling or buying, mark to market ceases to have the same information content. Why are buyers no longer there? Could be that liquidity disappears. Pseudo-arbitrageurs, agents who bring things back to fundamentals don't have the firing power to deploy their money. Two sides, some information still there. Loss of trust may simply reflect vagaries of today's supply and demand. Mark-to-market requirement and regulatory capital buffer: If I'm holding an asset, will hold for 5 years. Mortgages will be paying out over time. Some will go into default, some will be prepaid, etc. I'm not going to resell. Meanwhile, if there is an increase in the default rate, the market value of this goes down. I may not be in compliance with my capital requirements today but over time I may be fine on a cash-flow basis. Somewhat compelling argument--forced to sell it because of the capital requirement, though I wouldn't ordinarily sell it. When creating a security, you have a choice to place it on the banking book or the trading book. Trading book attracts less capital but forces you to mark to market on a daily basis. Perhaps people have been placing into the trading book when the sky was blue, price high, prices almost for everything, a lot of things whose prices disappeared when liquidity was withdrawn. Traded loan on trading book; but if you plan to hold it to maturity, place it on your banking book, in which case you get a different capital and a different recognition of a value. Unless impairment, value of 100. Choices that were made through 2006-2007 was to place in trading book--less capital, abundant liquidity, blue sky days. |
| 47:53 | Too big to fail. Enlightened self-interest, or maybe just self-interest. Some have argued that too much risk was taken because at the end of the day, "too big to fail" or U.S. Central Bank behavior would bail out these bad decisions. In the United States we have banks that are regulated by the Federal Deposit Insurance Corporation (FDIC). Also shadow banking system, investment banks not guaranteed, but it turned out that they kind of were--erratic decision: Lehman Brothers died, Bear Sterns sold at discount, Merrill Lynch sold to Bank of America, AIG made whole all the way through. How much was there a socialization of losses and a private set of gains? Fannie Mae and Freddie Mac turned out to be a catastrophic institutional structure that turned out to be implicitly guaranteed; many investment banks were not guaranteed at all. Was that rational? Instead of thinking of an institution in the abstract, think of the decision-makers. Perhaps institution might survive, but the individuals who made those bad choices don't. Deterrence from rolling the dice. Could argue it goes the other way: a lot of traders made a great deal of money along the way, and though some CEOs were both humiliated or financially destroyed, others turned out fine. CEO of AIG announced his resignation today, saying he was leaving the most horrible job in the world. Before the event, cannot know if you are going to be one of the survivors or not. If you are rescued by the government you have lost control. |
| 51:58 | Systemic risk. There is a negative externality, so when you take a risk you are careful, but you are not as careful as you would be if you had to bear the other social costs of the other firms that will be called into question. Back to March of 2008, turning point when U.S. Federal Reserve became active in the financial system to a degree that had been unimaginable previously. Weekend in March, Bear Sterns informed Fed and Treasury that it was not going to be able to meet its obligations. Failure of democracy that Fed and Treasury never explained the urgency other than saying it was urgent and that it had to be done. Meltzer podcast, few cases where firms were allowed to fail, failure of incentives. Claim was that if they had failed so much would have cascaded down. A few months later they let Lehman Brothers fail, which people have debated. We now have information about their assets--they've gone through bankruptcy. Claim was that if Bear Sterns had been allowed to go through that, whole system would have frozen up. But whole system froze up anyway. Was that Bear Sterns an unsustainable event? What if it had been allowed to go through bankruptcy? Intense debates. Should we let the market work out its own problems or should we intervene? Records of different central bankers, they were more averse to interventions. Some central banks said rescue of Bear Sterns was a mistake. When Lehman came around, people who favored the open market had the upper hand. When they saw what happened, the central banks said that's it, no more. We would have had a preview of Lehman in March. John Taylor has a claim that the spike in the spreads occurred after the Lehman failure, not concurrent; and it was really the TARP bailout that spiked interest rates. Don't agree; there was real panic after Lehman was allowed to fail. Noise prompting depositors to take their money out. Pure speculation, counterfactual. Regulators in the dark--face their own set of incentives. In the United States, Ben Bernanke, an expert on the Great Depression was eager to not have another one under his watch. |
| 57:09 | European perspective: tendency to blame much of the crisis on various U.S. policy decisions behind the housing crisis. In 1990s, and can even trace it back to the 1930s under Roosevelt, United States eagerly subsidized price of housing, inflating housing prices. Blames tax policy, Fannie and Freddie, but was it large enough? Those sympathetic to that argument often ignore similar housing price appreciations in Europe: Spain, Ireland, South Africa, Australia around the world. John Taylor argument: excess of liquidity searching for yield as central banks pushed out money. When something of the recent magnitude occurs, there is no single cause. Perfect storm, confluence. China prompts them to invest in Treasuries, depressing the yields and Fannie and Freddie, pumping them into housing in particular. Why housing? Environment with pretty low return on lots of assets in 2006, for many investors the most solid thing to invest in was houses. Home refinancing in the United States; houses looked at not as place to live but a place to invest. Magnitude and complexity and leverage of the instruments attached to mortgage securities. Government aiding and abetting increase in housing, but not the only cause. Money chasing places to go. But why Ireland and Spain and not England and France? Have to look at data; suggest looking at availability of mortgages. |
Imagine you are at a magic show. The magician takes an ordinary pack of 52 playing cards, and gives it to a man in the audience to shuffle. He then asks a volunteer to think of a card. "The ace of spades," she replies. 
Posts
