Category: Education

Presentation Description

No description available.


Presentation Transcript


Behavioral Finance: Implications for Insurance and Reinsurance Markets CAS Annual Meeting - November 13, 2007 Don Mango, Guy Carpenter Richard Goldfarb, Benfield Advisory

Should Be Our Only Slide –Council of Insurance Agents & Brokers: 

Should Be Our Only Slide –Council of Insurance Agents & Brokers NEWS RELEASE – Monday, Oct. 22, 2007 Third Quarter Market Survey Shows Easing Conditions for Coastal Property WASHINGTON -- The soft market conditions that have affected most commercial property/casualty lines for more than a year are beginning to spread to coastal properties with wind storm exposures…Commercial agents and brokers responding to the survey said coastal properties with wind exposure are still the hardest risks to insure, but market conditions are beginning to ease, and more capacity is available. These were among the comments to open-ended questions in the survey, all from brokers based in the Southeast: "Most wind accounts are seeing 20-30 percent decreases year-over-year on the same limits." "Rates are coming down for wind insurance. We thought we might see a change entering hurricane season, however carriers are still quoting aggressively. No significant changes, however, in deductibles." "We're in Florida. The marketplace has gotten very competitive for wind, especially with the smaller regional carriers."

Should Be Our Only Slide – Andrew Lo: 

Should Be Our Only Slide – Andrew Lo One anecdotal piece of evidence is provided by the collapse of fixed-income relative-value hedge funds in 1998 such as Long-Term Capital Management (LTCM)… ...In retrospect, even the most ardent critics of LTCM and other fixed-income relative-value investors now acknowledge that their spread positions were quite rational, and that their demise was largely due to an industry-wide underappreciation of the commonality of their positions and the degree of leverage being applied across the many hedge funds, investment banks, and proprietary trading groups engaged in these types of spread trades. This suggests that the forces of irrationality—investors flocking to safety and liquidity in the aftermath of the Russian default in August 1998—were stronger, at least for several months, than the forces of rationality.

Should Be Our Only Slide – Pick a Date: 

Should Be Our Only Slide – Pick a Date Friday November 9, 10:24 am ET By Joe Bel Bruno, AP Business Writer Stocks Fall Again, Investors on Edge As Wachovia Takes Writedown, Economic Worries Persist NEW YORK (AP) -- Wall Street extended its slide in early trading Friday after Wachovia Corp. warned it will take quarterly loan losses due to its debt portfolio, raising investor concerns that the credit slump shows no sign of abating… Investors also were rattled by speculation that Barclays PLC was about to announce a $10 billion writedown… Further worries about the continuing credit market slump kept investors on edge a day after Federal Reserve Chairman Ben Bernanke said he expects the economy to "slow noticeably" this quarter. His comments added to the week's declines -- contributing to a slide this week on concerns about continuing credit woes, a weakening dollar and rising oil prices.


It’s the Rule not the Exception Know your own limitations Know the limitations of your colleagues and customers Modify your actuarial communications accordingly Help management make better decisions under uncertainty

What We Will Cover: 

What We Will Cover Overview of Behavioral Finance Background: Modern Financial Theory Behavioral Finance – Key Concepts and Examples Behavioral Implications for Insurance & Reinsurance Markets Traditional Actuarial Practice: Actuarial Analysis and Product Pricing Emerging Actuarial Practice: Internal Risk Models (ERM) Emerging Regulatory Practice: Fair Value Measurement for Financial Reporting and Solvency Assessment The Market Q&A

Background on Modern Financial Theory: 

Background on Modern Financial Theory Portfolio Theory Investors rationally evaluate risk and return trade-offs of potential investments Identify an “optimal” portfolio of risky assets that suits their risk tolerance, objectives and constraints. Capital Asset Pricing Model All investors follow the approach described above and can borrow/lend at risk free rate Results in a common “optimal” portfolio of risky assets for all investors Leads to an equilibrium price (or expected rate of return) for risky assets Dependent upon definitions of “risk” – e.g. standard deviation of returns Market Efficiency Underlying assumption that prices are generally “right” due to an absence of arbitrage Option Pricing When portfolios can be continuously rebalanced and risk continuously/perfectly hedged, it is possible to use relative pricing to value risky cash flows Black-Scholes option pricing is one important example Dependent on the absence of arbitrage opportunities

Behavioral Finance – Overview: 

Behavioral Finance – Overview Behavioral Finance – Argues that the methods people actually use to evaluate risk and make decisions under uncertainty differ materially from what underlies modern financial theory. Models and parameters are estimated with tremendous uncertainty Investors unable to articulate their Utility Functions – at least not consistently across different situations Real decisions rarely reflect simple maximization of expected utility Critical Foundation – Experimental work by Kahneman and Tversky, which led to a Nobel Prize in Economics for Kahneman in 2002. Three Themes of the Behavioral Finance Research Common Processing Errors Associated with Evaluating Risk Challenges for Decision Making Under Uncertainty Limits to Arbitrage – Impacts Market Efficiency

Behavioral Finance – Common Processing Errors (1 of 3): 

Behavioral Finance – Common Processing Errors (1 of 3) Overconfidence People overestimate their knowledge, underestimate risks, and exaggerate their ability to control events Example: 130 actuaries were asked to estimate what Wal-Mart’s total revenue was in 1999 and provide a 90% confidence interval for their estimate Reasonable to assume that 90% of the ranges would include the actual amount Only 28% of the ranges included the actual amount Anchoring People tend to anchor on existing estimates and do not adjust their estimates adequately to reflect new information. Often the anchoring phenomenon is significant even when the initial estimate is known to be irrelevant, baseless or highly uncertain

Behavioral Finance – Common Processing Errors (2 of 3): 

Behavioral Finance – Common Processing Errors (2 of 3) Representativeness The subjective probability of an event, or a sample, is determined by the degree to which it: (i) is similar in essential characteristics to its parent population; and (ii) reflects the salient features of the process by which it is generated. In particular, since sample size does not represent any property of the population, it is expected to have little or no effect on judgment of likelihood. Example: Actuaries were given the following description of a woman: Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice and she participated in antinuclear demonstrations. When asked to rank the likelihood that various statements about Linda were true, 66% thought it was more likely that she was a bank teller and a feminist rather than just a bank teller. Law of Small Numbers (Gambler’s Fallacy) When results are above the long term mean, many people predict that future results will be below the mean. In reality, regression to the mean only predicts that future results will be closer to the mean, not below the mean. People mistakenly expect properties of large samples to be repeated in small samples

Behavioral Finance – Common Processing Errors (3 of 3): 

Behavioral Finance – Common Processing Errors (3 of 3) Confirmation Bias When evaluating the quality of estimates people often give too much weight to confirming evidence. Availability Bias When evaluating the likelihood of events, people tend to give too much weight to facts that are familiar to them. Example – People tend to overstate the probability of homicides relative to strokes given the attention paid to homicides in the media. Hindsight Bias People often have a selective memory of success and take credit for luck.

Behavioral Finance – Decision Making Under Uncertainty (1 of 2): 

Behavioral Finance – Decision Making Under Uncertainty (1 of 2) Framing Decisions under uncertainty are often impacted by how the situation is framed. Mental Accounting A form of framing occurs when people mentally, and arbitrarily, bifurcate decisions Example – Tendency to become more willing to take risks with “the house’s money”. Regret Avoidance People tend to make choices that minimize the chance of regret. Causes “herd” behavior.

Behavioral Finance – Decision Making Under Uncertainty (2 of 2): 

Behavioral Finance – Decision Making Under Uncertainty (2 of 2) Myopic Loss Aversion People generally like to avoid losses and exhibit loss aversion However, they also tend to evaluate risks in isolation rather than in an aggregate context. This leads to excessive risk aversion – sometimes called myopic loss aversion. Suggests why people are drawn to loss damage waivers when they rent a car, buying warranties on products and purchasing low deductible/low limit insurance policies. Prospect Theory (Kahneman & Tversky) Many of the above phenomena can be explained by Prospect Theory Recall that conventional wisdom is that people make decisions under uncertainty by maximizing expected utility, with utility measured with respect to total wealth Prospect theory instead suggests that people evaluate gains and losses relative to current wealth People will be more risk averse with respect to gains and risk-seeking with respect to losses

Impact of Framing on Risk Aversion: 

Impact of Framing on Risk Aversion Two Groups Given Identical Problem Framed Differently Both required a choice between a risky and risk-free outcome Mathematically the problems were identical One framed as a potential for gain, the other as a potential for loss Survey Results Group 1 favored the Risk-Free option Group 2 favored the Risky option

Behavioral Finance – Limits to Arbitrage (1 of 2): 

Behavioral Finance – Limits to Arbitrage (1 of 2) Key Question: Do behavioral biases and the failure to use conventional expected utility actually impact market prices for risky assets? Conventional wisdom: when individuals make “errors” professional arbitrageurs will step in to exploit those errors, thereby driving market prices back to their “proper” level. This assumes there are no limits to arbitrage Fundamental Risk Arbitrageurs who believe they have identified a mispriced asset can in fact be mistaken Exploiting apparent arbitrage opportunities is therefore inherently risky Some marginal mispricing may therefore persist Noise Trader Risk Noise traders are “irrational” investors who do not trade based on fundamental values Mispricings that result may be exploitable, however by definition they are irrational and can get more irrational, causing prices to further deviate from fundamental values Someone seeking to exploit the mispricing faces the risk that their losses pile up over short horizons Example: AQR almost went out of business betting against the internet/technology bubble in the late 1990’s (luckily they survived though and now manage over $30 billion!)

Behavioral Finance – Limits to Arbitrage (2 of 2): 

Behavioral Finance – Limits to Arbitrage (2 of 2) Implementation Costs We often ignore the transaction costs, frictional costs, short selling costs and monitoring costs associated with exploiting market mispricings Since these costs are real, mispricing can persist Example: 3Com’s Spinoff of Palm in 2000 For an extended period of time, Palm’s market capitalization significantly exceeded 3Com’s market capitalization, even though 3Com owned 94% of Palm. Model Risk Valuation models can indeed be wrong (surprise!) Short Horizons & Liquidation Risk When investors such as professional portfolio managers or hedge funds are faced with short horizons, liquidation risk can be substantial Examples: LTCM, Blowup of Statistical Arbitrage Funds in August 2007, Morgan Stanley’s 4th Quarter 2007 Super Senior CDO Losses

Behavioral Finance and Actuaries: 

Behavioral Finance and Actuaries Probability judgments Curse of knowledge Law of small numbers Heuristics – rules of thumb “Quasi-Bayesian”— people misspecify a set of hypotheses, or encode new evidence incorrectly, but otherwise use Bayes’ rule. Job definition Unaware of how little others know Impacts communication Over-reliance on small samples Out of necessity E.g., high layer capital = 2 x Limit Weighting experience and exposure Large loss adjustment We got priors

Behavioral Finance and Actuaries: 

Behavioral Finance and Actuaries How Should These Insights Impact Existing and Emerging Practice? Traditional Actuarial Practice: Actuarial Analysis and Product Pricing Emerging Actuarial Practice: Internal Risk Models (ERM) Emerging Regulatory Practice: Fair Value Measurement for Financial Reporting and Solvency Assessment

Actuarial Analyses and Product Pricing (1 of 2): 

Actuarial Analyses and Product Pricing (1 of 2) Anchoring Impacts… Renewal pricing B-F Method Reserve reviews Increased Limits Factors Regret Avoidance Impacts… Actuarial Best Estimates Link Ratio selections Tail Factors B-F Seed ratios Overconfidence Impacts… Berquist-Sherman Method Excess LDF Selection Trend selection Tail Factors Reserve ranges What do we use? What did we use last year? What did I tell you last year? This is THE number Five (seven, nine) data points

Actuarial Analyses and Product Pricing (2 of 2): 

Actuarial Analyses and Product Pricing (2 of 2) Hindsight Bias Impacts… Trend selection Reserve reconciliation Representativeness & Law of Small Numbers Impact… Underwriting and risk selection Confirmation Bias Impacts… Actual vs. Expected Analysis Mental Accounting Impacts… Segmentation of Loss Triangles Status Quo Bias Impacts… Reserve reviews Experience vs. exposure rating weights Endowment Effect Impacts… Renewal business and retention strategies My (GL, WC, Umbrella) triangle has 30 AY’s Subject to nearly ALL behavioral issues

Internal Risk Models: 

Internal Risk Models Internal Risk Models Taking on Great Importance for Insurers Pressure from rating agencies, regulators, investors to demonstrate understanding of company risk profile Desire to assess capital adequacy and use capital efficiently Desire to allocate or attribute capital to specific activities, businesses, etc. Desire to evaluate risk-reward trade-offs explicitly and consistently Behavioral Finance Issues Impact All Aspects of These Models Inputs Outputs Validation Management Use of Model Results

Internal Risk Models – Inputs: 

Internal Risk Models – Inputs Challenges Limited and inconsistent data Judgment and expert opinion play critical roles Extrapolation from historical results, data and exposures Quantifying correlations across risks, across lines of business, etc. Behavioral Issues to Be Mindful Of... Anchoring Overconfidence Hindsight bias Regret avoidance

Internal Risk Models – Outputs: 

Internal Risk Models – Outputs Challenges Myriad of risk measures to choose from, yet no clearly superior risk measure for all applications Limited comparability Focus is often on extreme events Substantial measurement error of extreme percentiles Very sensitive to subtle and often arbitrary assumptions or modeling choices Difficult to interpret Behavioral Issues to Be Mindful Of... Overconfidence Anchoring Do people really make key decisions based on extremely unlikely events? Do you ever decide whether to take a plane, train or automobile based on the relative mortality rates from each? How do people really make decisions under uncertainty?

Internal Risk Models – Validation: 

Internal Risk Models – Validation Challenges Definition of “valid” Reliability “Prove it’s correct” “What’s the optimal capital level?” “What’s the proper capital allocation?” Behavioral Issues to Be Mindful Of... Loss aversion Anchoring (e.g. to AM Best’s BCAR) Overconfidence How do we know we’ve chosen the right copula? How do we know the method is reasonable? How do we know the results are realistic?

Internal Risk Models – Use Test: 

Internal Risk Models – Use Test Use Test = regulatory term from UK FSA In order to get “credit” for internal capital model, firm must demonstrate Management understands the model and can explain how it is used Models impact major decisions Models represent the official risk record of the organization (Goldman Sachs) Value Comes from Using the Models to Make Better Risk Decisions Consistency Transparency Objectivity Requires articulation of risk preferences, risk tolerances and risk appetite Unfortunately, these are ill-defined, malleable and context-sensitive

Internal Risk Models – Use Test (2 of 2): 

Internal Risk Models – Use Test (2 of 2) Recall all of the reasons why decisions under uncertainty are so difficult and are not made simply by maximizing expected utility: Framing Mental Accounting Regret Avoidance Myopic Loss Aversion Prospect Theory (Kahneman & Tversky) Formalized capital modeling amounts to unearthing a collection of inconsistent heuristics, biases and cognitive errors… …and trying to reach a compromise that “everyone can live with” Ricardo Rebonato’s Cautionary Note (in Plight of the Fortune Tellers, 2007): Our attempts at managing risk may be becoming more complex and cumbersome, but less effective.

In Search Of…the Insurance “Market”: 

In Search Of…the Insurance “Market” Market microstructure analysis slows down the movie The “market price” not even well defined in the capital market space Essentially = the most recent executed trade There have to be individuals on both sides of a trade Liquidity requires differences of opinion AND losers (Re)insurance underwriter opinion matters These are the components driving “insurance liquidity” (capacity, availability): Loss cost estimates Individual company results for the line Market signals (e.g., CIAB survey) Customer expectations (anchored on the expiring premium)

Fair Value Measurement for Financial Reporting and Solvency Assessment: 

Fair Value Measurement for Financial Reporting and Solvency Assessment Fair Value Measurement of Insurance Liabilities Reflects the price that would have to be paid to transfer the liabilities, in an arms length transaction among willing counterparties. Two Emerging Uses Financial Reporting – Both GAAP and IASB accounting standards are moving rapidly towards a fair value paradigm. Solvency II – Europe’s emerging regulatory capital standards for insurers requires capital to reflect the potential adverse change in fair values of net assets and liabilities. Practical Challenges Fair value measurements often rely upon relative valuation methods, which require some degree of market efficiency. Even for relatively liquid financial instruments behaviorists caution against always trusting market prices Consider the current inability of banks and hedge funds to properly value credit derivatives and other structured financial products. Insurance liabilities have less price transparency and are therefore even harder to value on a relative basis – requires broad agreement on methods for pricing insurance risk. Some have advocated widespread adoption of the Cost of Capital Method, though significant implementation challenges have yet to be sorted out

The Changing Face of the Capital “Market” (1 of 2): 

The Changing Face of the Capital “Market” (1 of 2) “U.S. Investors Face An Age of Murky Pricing,” WSJ Cover Story Oct 12, 2007 Today, "way less than half" of all securities trade on exchanges with readily available price information, according to Goldman Sachs Group Inc. analyst Daniel Harris. More and more securities are priced by dealers who don't publish quotes. Billionaire investor Warren Buffett advocates more transparency in pricing. "Some marks can be pretty imaginative, " he says. "They call it 'marking to market,' but it's really marking to myth." He says that before funds publish financial statements, they should sell 5% of hard-to-value positions to gauge values. Michael Vranos, a veteran mortgage-bond trader, recently told investors in his large hedge-fund company, Ellington Management Group, that he was suspending investor redemptions at the end of September because he couldn't figure out values for some of the fund's mortgage-related investments. "There is no way to determine [values] that would be simultaneously fair both to investors redeeming from these funds and to investors remaining in the funds."

The Changing Face of the Capital “Market” (2 of 2): 

The Changing Face of the Capital “Market” (2 of 2) “U.S. Investors Face An Age of Murky Pricing,” WSJ Cover Story Oct 12, 2007 Since the recent market turmoil, some traders seem to be trying to set prices simply by offering them for sale, according to some investors who say they've talked to traders. The position is then marked at the "offer” price, these investors say. That's akin to a homeowner valuing a house based on how much he wants for it -- not how much a buyer is willing to pay. In August, Thornburg Mortgage Inc. also discovered how unstable pricing had become when it had to liquidate $22 billion worth of high-quality mortgage securities, at a loss of $1.1 billion, to reduce its short-term debt. Larry Goldstone, the mortgage lender's president and chief operating officer, complains that some Wall Street dealers that were financing his company's activities sold his firm's mortgage securities at fire-sale prices to make sure they were repaid. In some cases, he says, dealers sold AAA-rated mortgage-backed securities at a much lower price than he was able to get for comparable securities at the same time, worsening his company's losses by tens of millions of dollars. "It was panic," he says. "Completely irrational."

Last Quote: 

Last Quote Todd Bault Most of the time the capital markets act like there is no parameter risk In times of crisis they act like there is only parameter risk Thank You Q & A

authorStream Live Help