Sticking with the December 6th anniversary theme, today’s book is Financial Risk Management: A Simple Introduction by K. H. Erickson, 2014. It’s an exciting little text on investment risk management techniques that are near and dear to my heart, as I enjoyed managing a significant portfolio for nearly 12 years. But, this is an amazing body of work as it provides an analysis focused exclusively on the 1994 Orange County Investment Pool implosion.
Maybe the best way to explain this book is to provide you with its Introduction:
Individuals, businesses, corporations, and governments are always searching for profitable investment opportunities which can offer an increased return. But the potential for a greater return will typically go hand in hand with greater risk, and there’s a danger than an investment strategy can backfire and end up costing more than it creates. Risk can come in many forms and while much risk can be avoided with well researched investments, or eliminated with a diversified portfolio, a degree of unavoidable market risk will always remain and therefore effective financial risk management is a central part of any investment strategy.
One of the most significant elements of market risk is interest rate risk, as changing yield rates can reduce the value of an asset or portfolio, and interest rate risk is a focus of this book. The field of financial risk management is explored in depth using theory, formulas, calculations, and examples, and then applied to the case study of the 1994 Orange County, California bankruptcy to examine whether the financial failure could have been avoided. Basic prior knowledge of derivatives and econometrics is assumed and used in the analysis.
Financial risk management involves first determining the risk exposure of an investment or portfolio, and this is explored using leverage, duration, modified duration, convexity, effective duration and effective convexity. Value at risk (VaR) is the next focus, and the three main variance-covariance, historical simulation, and Monte Carlo methods are explained and compared, along with the related square root of time rule. Detailed steps to calculate the variance-covariance, historical simulation, and Monte Carlo value at risk in Excel are provided. An exponentially weighted moving average (EWMA) is then introduced to predict factor change, interest rate or return volatility, and simple steps to calculate the EWMA and backtest the data for reliability in Excel are presented.
An extended empirical case study for Orange County’s bankruptcy in 1994 takes up the remainder of the book. First the history of how the situation came to pass is explained, with Orange County’s balance sheet, leverage, duration, effective duration, and modified duration examined to determine the extent of the county’s risk exposure, and assess whether the bankruptcy was inevitable or if alternatives were available. This discussion is then built upon with detailed value at risk and EWMA analysis as the potential for risk management is debated. The final section looks into theoretical hedging strategies for Orange County, examining a range of financial derivatives and how they may be used to hedge interest rate risk.
I hope that whets your appetite. It sure did mine. This short book covers all of what was introduced above. Let me provide Chapter 5.1, “Background to Orange County Failure,” and the author’s conclusion in Chapter 6.1, “Value at Risk for Orange County,” which may surprise you. As a bonus, I’ll also provide a segment of Chapter 7, “Hedging Strategies for Orange County,” as that was my focus the second half of 1994, if only someone from the County’s management team had called for my advice (as my efforts to contact them were rebuffed).
Background to Orange County Failure
The 1994 bankruptcy of Orange County, California is an interesting real life case study with which to further investigate financial risk management. In December of 1994 Orange County filed for bankruptcy and ultimately reported a loss of $1.64 billion, at the time the largest municipal failure in US history. Although it has since been exceed in scale in the US by the municipal failures of Jefferson County, Alabama in 2011 and Detroit, Michigan in 2013, the Orange County case retains significant interest due to the cause of the bankruptcy, which was taking risky positions in financial markets without a risk management strategy which covered against interest rate risk.
Robert Citron, long-time treasurer and tax collector of Orange County in 1994, was in a desperate position after local government’s tax allocations were cut by the state, and he needed a way to raise income for the county without raising taxes which would alienate the local electorate. Citron embarked on a risky strategy to raise income by borrowing heavily to invest in bond securities which would gain value if interest rates fell, as US interest rates had been following a general downward trend (The Public Policy Institute of California, 1998). This strategy also took advantage of natural uneven bond convexity, noted earlier, where increases in yields reduce price and value less than an equivalent yield decrease raises price.
The chosen investment portfolio for the Orange County Investment Pool (OCIP) mostly consisted of fixed income securities and structured notes, with additional funds obtained with the issuing of reverse repurchase agreements, a debt where a security is bought with the agreement to sell it on later for a higher price. This reverse repurchase agreements debt ensured a highly leveraged and risky position for Orange County’s books.
An examination of empirical evidence on historical portfolio performance explains Citron’s choice of assets. Asness (1996) notes that a levered portfolio of stocks and bonds outperformed an unlevered portfolio of 100% equities over the years 1926-1993. Overall a levered portfolio offered higher total returns than the unlevered comparison, with the same long-term consistency, and comparable standard deviation and worst case scenario outcomes. Based on this information Bob Citron’s choice of investment appeared wise, and it was successful at first as US interest rates fell steadily over the years 1989-93 to see the portfolio rise in value.
February 1994 saw the beginning of a series of sharp rises in US interest rates as the Federal Reserve looked to tighten credit to reduce the threat of inflation and prevent the US economy from overheating. The constant maturity treasury rate rose from 3.45% in February 1994 to 7.14% in December 1994 (from 3.61% a year before in December 1993). These yield rises caused Orange County’s portfolio to plummet in value, and some predicted Citron would lead the county to bankruptcy. John Moorlach was one dissenting voice and he ran for public office against Citron in the June 1994 Orange County Treasurer local election, opposing the reckless investments Citron had made with billions of dollars of public money. But Bob Citron won the public vote and remained in control of the Orange County Investment Pool for the rest of 1994.
In an attempt to reclaim funds lost from rising interest rates and secure the high returns required for his county Bob Citron went on to borrow further, doubling down on the investment strategy of betting on falling interest rates and a yield trend reversal. But the reversal didn’t come in time and when Citron couldn’t keep up with the margin payments on the debt funds borrowed in the repo market, due to low returns from a bad investment strategy, the game was up. Orange County was unable to sell portfolio assets seen by other investors as too risky, and when creditors moved in to seize assets as collateral in December 1994 the county declared bankruptcy.
Value at Risk for Orange County — Conclusion
Putting all of the analysis together it seems that all three of the value at risk methods, variance covariance, historical simulation, and Monte Carlo simulation, have their weaknesses. None can be used to accurately predict what actually happened to Orange County in 1994, leaving little confidence that they would be able to predict what would happen in the future to allow risk management. And the square root of time rule which allows monthly interest rate changes to be aggregated for longer value at risk periods appears unusable for Orange County, as interest rates appear to exhibit stationary characteristics which violate the rule’s key assumptions. This is another obstacle preventing successful long-run financial risk management. It seems that the value at risk measure wouldn’t have helped Orange County much in the 1994 period where they faced financial problems, and other tools are there required.
Hedging Strategies for Orange County
Orange County could have invested their resources into hedging interest rate risk in late 1993 or early 1994, before the damaging yield rises removed the option. But although methods were available at that time to hedge interest rate risk and offset possible financial loss with opposite and equivalent investment positions, as Orange County ultimately required in 1994, investment of resources in such tools would by definition take resources away from those assets which generated profits in the event that interest rates fall.
This e-mail has been sent by California State Senator John M. W. Moorlach, 37th District. If you no longer wish to subscribe, just let me know by responding with a request to do so.