Credit Risk

Credit risk is the chance that the issuer of a security will be either unable or unwilling to repay its debts. The full benefits of owning investments with low credit risk are derived if these investments can be held to maturity. Under such circumstances, owners are ensured that they will receive back, not only the principal of their investments but also, the full investment yield. On these grounds, the OCIP portfolio was rock-solid safe because the assets Citron purchased were restricted to securities issued by the U.S. government, U.S. government agencies (e.g., Import-Export Bank), and high-quality companies (e.g., Bank of America and Ford Motor Company), all of which had a very low chance of declaring bankruptcy.

Market Risk

Market risk is the variation in the price of an asset (or liability) due to changes in financial variables (e.g., interest rates, exchange rates, stock prices, and/or commodity prices). There is an inverse relationship between the price of a fixed-income security and its yield, and the longer an asset’s maturity, the greater is its potential price variation for each percentage change in yield. Market risk has nothing to do with whether borrowers can or will repay their debts, but rather it captures the possible change in an asset’s value if it is sold prior to maturity. The issuers of this debt (i.e., the borrowers) are under no obligation to repay their liabilities before maturity; so, if investors want their funds back early, they must sell their securities for what the market bears. As interest rates rise, the prices of fixed-income securities fall, and these investors incur capital losses. Likewise, if interest rates fall, the prices of these securities rise and investors earn capital gains.

Table 5.3 shows how the prices of four zero-coupon securities with different maturities vary when interest rates rise from 3.5 to 7 percent (as they did during 1994). Notice that the one-year bill changes in value by only –3.3 percent, but the five-, ten-, and twenty-year securities change by about –15, –28, and –49 percent, respectively. The longer the maturity, the greater is the decline in the price of a fixed-rate note when interest rates rise. Portfolio managers, who are caught in a position where they have to sell fixed-income assets as interest rates rise, could lose a considerable amount of capital.

Table 5.3: Change in Price of Zero Coupon Securities with One-, Five-, Ten-, and Twenty-Year Maturities When Yields Rise from 3.5 to 7.0 Percent (Face Value = $1,000)

Orange County faced substantially more market risk than credit risk because OCIP was invested mainly in medium-term securities, and it was highly leveraged (i.e., large amounts were borrowed to finance the purchase of securities). As a result, the value of the OCIP portfolio varied significantly with changes in market yields.

Liquidity Risk

Liquidity is the ability to turn an asset into cash quickly and without substantial loss of value. Liquidity risk for a fund manager is the chance of not having sufficient cash on hand to meet demand. With regard to liquidity risk, investment funds are like banks. If a large number of customers suddenly demand their deposits, no bank or fund would have enough cash on hand to meet that demand. This means, even if a fund’s assets were safe from credit risk, it could fail for lack of liquidity.

Liquidity risk can be tied directly to market risk because if interest rates rise and the value of a portfolio falls, investors could decide to withdraw their funds, thereby forcing portfolio managers to sell investment assets at their low current-market prices, resulting in losses that could threaten the fund’s solvency. By outward appearances, OCIP had more than enough cash assets to meet normal demands by customers, brokers, dealers, and creditors. Nevertheless, in a crisis, cash demands become amplified, magnifying liquidity risk.

Although Citron and the Orange County board believed OCIP was protected from credit risk due to the constraints on the types of investments Citron could make, it was vulnerable to the confluence of all three risks. An overview of OCIP’s assets and funding sources demonstrates this vulnerability.

Structured Notes

Almost 40 percent of the OCIP portfolio was invested in structured notes.125 Structured notes are customized investments with periodic payments that depend on future changes in market variables, such as interest rates, exchange rates, stock prices, and commodity prices. Brokers and dealers like structured notes because they are customized and earn greater commissions and fees than straight bond or note issues; investors like them because they offer opportunities not readily available in the market. Robert Citron responded to the incentives of structured notes by investing almost $8 billion in these securities. Such investments would have posed no problems had Citron fully understood their risks, but therein was the problem.

Despite a lack of understanding, Citron invested heavily in four types of structured notes with varying degrees of risk. Of the four types, inverse floating-rate notes were the largest category and a major source of OCIP’s losses.

Inverse Floating-Rate Notes

Most of OCIP’s structured notes and more than 26 percent of its total assets were placed in inverse floating-rate notes (also called inverse floaters or reverse floaters). As the name suggests, inverse floaters are financial instruments whose returns vary inversely with market interest rates, rather than directly, as would be the case with floating-rate notes. When market rates rise, inverse floaters earn lower returns; when rates fall, the returns on inverse floaters rise.

An example demonstrates how the return on an inverse floater varies with market rates. Suppose on January 1, 1993 a portfolio manager purchased an inverse floater that earned an annual return equal to 7 percent minus the six-month LIBOR. If the six-month LIBOR rate on January 1 were 4 percent, then the inverse floater would earn a 3 percent annual return (i.e., 7% – LIBOR of 4% = 3%), which means it would earn just 1.5 percent for the six months from January to June (i.e., 6/12 × 3% = 1.5%). On July 1, six months later, the return would change. If LIBOR rose to 5 percent, the annual return on the inverse floater for the following six months would fall from a 3 percent to 2 percent (i.e., 7% – LIBOR of 5% = 2%), earning 1 percent from July to December; if interest rates fell to 3 percent, the annual return on the floater would rise to 4 percent (i.e., 7% – LIBOR of 3% = 4%), earning 2 percent for the second half of the year.

Robert Citron was betting on falling U.S. interest rates; so, it is easy to understand why inverse floaters were appealing to him. Inverse floaters also could be tailored to fit his risk specifications, and Citron’s inverse floaters were often structured to carry an extra wallop when interest rates changed. Some of his super-charged inverse floaters were designed to pay off at twice the change in LIBOR (e.g., 8% – 2 × LIBOR, rather than 8% – 1 × LIBOR), but there was a flip side. As interest rates rose, the earnings on these structured notes fell at twice the rate of a normal inverse floater.

Fixed-Income Securities

Fixed-income securities were the meat and potatoes of the OCIP portfolio. Constituting nearly 58 percent of the pool’s assets, these securities paid either a fixed rate of interest or a rate that varied according to a predetermined schedule. They carried virtually no credit risk because the issuers were high-quality borrowers, such as the U.S. government, U.S. agencies, and first-class corporations. The average maturity of OCIP securities was under three years, but because these investments were leveraged, the impact of changing market rates on their values was amplified. As U.S. interest rates rose, the value of these securities plunged.

OCIP’s Funding Sources

In addition to concern about the composition of OCIP’s assets, there was also concern about the structure of its funding sources. Investors contributed only $7.6 billion to OCIP, but, as Tables 5.4 and 5.5 show, the pool held assets equal to $20.5 billion. The difference between OCIP’s total assets and investors’ funds (i.e., almost $13 billion) was the amount Citron borrowed to finance additional assets. The ratio of OCIP’s total assets to its equity, 2.7, was the leverage of OCIP’s portfolio (see Table 5.5).

Table 5.5: Orange County Investment Pool Portfolio: December 1, 1994 (Millions of Dollars)

Source: See Table 5.4.

Reverse Repurchase Agreements

To borrow such vast amounts, Citron made extensive use of reverse repurchase agreements (reverse repos, for short). To understand what went wrong at OCIP, it is crucial to understand what reverse repos are and how they can be used to leverage an investment portfolio. It is helpful to start with a short explanation of repurchase agreements (i.e., repos).126

A repo agreement is the spot sale of an asset with a simultaneous agreement to buy it back at a fixed price in the future. Usually, the asset sold and repurchased is a fixed-income security, and when the term repurchase agreement is used, securities dealers are the borrowers (i.e., the security sellers) and their customers are the lenders.

On one hand, repos are just like collateralized loans, but on the other hand, they are like forward contracts. Repos are like collateralized loans because borrowers pledge assets (e.g., fixed-income securities) that back their promises to repay principal and interest in the future. For the counterparty who sells a bond and agrees to repurchase it in the future (i.e., the borrower), a repo transfers possession of the security in return for hard cash. When the security is repurchased, the difference between the original sale price and the higher repurchase price is the effective interest paid.

When repo borrowers repay their debts (which is normally the case), they become owners of the pledged collateral and receive all the interest and capital gains accruing to these assets. But if repo borrowers are unable or unwilling to meet their contractual obligations, lenders take possession of the pledged collateral, and they can sell it to extinguish the debts. To lenders in a repo deal, credit risk depends more on the quality of the collateral than it does on the quality of the borrowers.

Repos are also like forward contracts because the repurchase prices of the pledged assets are identical to forward prices. The difference between the spot price (the current sale price of the security) and the forward (repurchase) price is like the capital gain earned on a derivative instrument. To borrowers, repos are like long forward contracts because they are agreeing to purchase assets in the future at prices agreed upon now. To the lenders, repos are like short forward contracts because they are agreeing to sell assets in the future at prices agreed upon now.

Figure 5.1 summarizes the cash flows of a repo agreement. Suppose a dealer enters into a one-year repo in which he pledges a fixed-income security in return for $100 million (see left side of Figure 5.1).127 At maturity (i.e., one year later, in this case), the dealer would repurchase the security for its principal plus the repo interest rate. Assuming the repo rate was 3 percent, the dealer would pay $103 million (see right side of Figure 5.1).

A reverse repo, which is how Citron borrowed most of his funds, is just like a repo agreement, except the roles of borrower and lender are reversed (see Figure 5.2). In a reverse repo deal, securities dealers are the lenders, who purchase securities from customers and agree to sell them back at a higher price in the future. Customers are the borrowers.

OCIP’s Rising Returns: Effects of Falling Interest Rates

Even a cursory look at Figure 5.3 shows why Citron, at one time, was considered a financial genius. From 1989 to 1992, U.S. Treasury security yields fell steadily, with only minor interruptions. For example, the one-year Treasury bill rate fell from (about) 9.0 to 3.7 percent, and the five-year Treasury note rate slipped from (about) 9 to 6 percent. As a result, Citron’s heavy concentration of fixed-income securities was rewarded handsomely with capital gains, and his inverse floaters earned increasingly higher returns.

OCIP was able to profit from these falling interest rates. In 1994, OCIP’s leverage factor was approximately 3:1 (see Table 5.5), which means that for each $100 million worth of securities purchased with investor funds, an additional $200 million worth of securities were purchased with borrowed funds (e.g., via reverse repo agreements). Suppose that, on the $200 million of borrowed funds, OCIP paid an annual (reverse) repo rate of 3 percent, and on the $200 million worth of additional five-year government bonds, it earned 3.5 percent. At the end of the year, the counterparties (e.g., Merrill Lynch) would expect OCIP to repay the principal of $200 million principal on the reverse repo plus $6 million of interest (i.e., 3% of $200 million).

Robert Citron was betting that interest rates would fall, and if they did, he positioned OCIP to earn returns that were even more attractive because, in addition to the interest income, his securities would earn capital gains. Suppose at the end of the year, interest rates declined from 3.5 to 2 percent. OCIP’s original $100 million security would now be worth $107.07 million, a capital gain of $7.07 million.130 As well, the $200 million of securities purchased with borrowed funds (i.e., the reverse repo funds) would now be worth $214.14 million.131 With these funds, OCIP could repay its reverse repo debt amounting to $206 million (i.e., $200 million of principal plus $6 million of interest). As Table 5.7 shows, the effect of these capital gains raises OCIP’s total return on its $100 million of equity from 3.5 percent, which it would have earned with no leverage, 25.71 percent!

Table 5.7: Results of Leveraging Reverse Repos to Enhance Investment Returns: An Example Assuming Falling Interest Yields

In addition to the capital gains earned from falling interest rates, Citron also received another source of profits on his reverse repo agreements because from 1989 to 1992, the yield spread between medium-term and short-term interest rates rose. For example, Figure 5.3 shows that the spread between five-year U.S. interest rates and one-year rates grew from approximately 0 percent to more than 2 percent. As a result, Citron profited by borrowing, via reverse repos, from the short-term portion of the yield curve and investing in assets on the medium-term and long-term portions.

OCIP’s Return Stabilizes: 1993

During 1993, U.S. interest rates stopped their dramatic fall and stabilized. One-year rates and five-year rates hovered around 3.4 and 5.1 percent, respectively (see Figure 5.3). As a result, Citron was deprived of the capital gains he had been earning on OCIP’s fixed-income securities, and his inverse floaters no longer earned increasing returns. Profits fell, as well, because the spread between medium-term rates and short-term rates decreased, causing the net return on Citron’s mountain of reverse repo-financed securities to decline. Figure 5.3 shows that that spread between five-year and one-year Treasury bills yields fell by nearly 1 percent. The combined effect of stabilized interest rates and declining interest-rate spreads caused OCIP’s profits to fall, but they remained relatively healthy—especially compared to other funds. As a result, Citron was still considered, by many, to be a cut above the typical portfolio manager.

OCIP’s Returns Plummet: 1994—Effects of Rising Interest Rates

In the end, Citron’s bet on falling interest rates proved disastrously wrong. U.S. interest rates began to rise in February 1994, when the Federal Reserve tightened credit to reduce the threat of inflation and prevent the U.S. economy from overheating. During the span of ten months, the Fed raised U.S. interest rates six times, causing the yield on one-year U.S. Treasury bills to rise from 3.54% to 7.14 percent—more than double! As a result, OCIP’s heavily packed portfolio of fixed-income securities and inverse floaters began to spill red ink. Fixed-income securities incurred capital losses, and the spread between medium-term and short-term securities was rapidly approaching zero (see Figure 5.3).

The effect of rising interest rates combined with leverage caused returns to plummet. At the end of 1994, interest rates rose from about 3.5 to 7 percent. As a result, securities worth $300 million (i.e., the original $100 million bought with investors’ funds plus the $200 million purchased with borrowed funds) at the beginning of the year were worth $256.95 million at the end of the year—a considerable difference.

At year’s end, OCIP would still have to repay $206 million in principal and interest on the reverse repo agreement; so, as Table 5.8 shows, the effects of these capital losses would turn the safe 3.5 percent return on equity, which OCIP would have earned without leverage, into a loss of 38.55 percent.

Table 5.8: Results of Leveraging Reverse Repos to Enhance Investment Returns: An Example Assuming Rising Interest Yields

OCIP’s profitability plunged in 1994 for another reason. Like so many other finance-related failures (e.g., Barings Bank), Robert Citron began using a doubling strategy. As interest rates rose and bond prices fell, Citron saw in each interest rate up-tick an opportunity to purchase (what he believed were) increasingly underpriced assets. To finance the purchases and profit from his expectations, he borrowed vast amounts, but interest rates continued to rise, causing his doubling strategy to lose even more than had already been lost.132

The dual effect of declining profitability and tumbling asset value attracted the attention of investors and creditors, who began to scrutinize carefully Citron’s risk levels and match them with expected rates of return. The reaction of many to the bad news was to exit from the fund (or, at least, try to exit) as quickly as possible.

The main cause of Orange County’s bankruptcy was the heavy concentration of structured notes and medium-term fixed-income securities in OCIP’s portfolio, the risks of which Citron did not fully understand. Citron made what amounted to an enormous one-sided bet that the U.S. yield curve would remain steeply upward sloping (i.e., short-term interest rates would remain considerably lower than medium-term and long-term rates) and that interest rates, on average, would fall. He lost his bet, on both accounts, because the U.S. yield curve flattened, and interest rates rose in 1994. Rising interest rates led to a domino effect among the major risks facing OCIP. As the market value of fixed-income securities plummeted (market risk), demands by investors for their funds back forced OCIP to liquidate a large portion of its portfolio (liquidity risk), thereby turning paper losses into actual losses (credit risk).

Market Risk Causes Liquidity Risk

Investors, already spooked by the effects that the Fed’s tight credit policy and rising interest rates were having on earnings and portfolio values, began fearing for the solvency of OCIP. Capital flight resulted, beginning in mid-November 1994, when the Irvine Ranch Water District decided to withdraw $100 million from the OCIP portfolio. Citron and his staff (mainly, Matthew Raabe) spent the rest of the month trying to convince investors not to withdraw their funds and urging creditors not to pull their credit lines.

To stem the capital outflows and prevent a liquidity crisis, Citron tried to freeze the fund for thirty days and applied a 20 percent haircut (i.e., reduction in portfolio value) to investors wishing to withdraw their funds. The policy seemed to be fair. After all, these investors enjoyed the better years when OCIP was earning above-average returns; so, why should they not have to bear the burdens during more difficult times? Even though it is questionable whether Citron’s 20 percent haircut would have survived a serious court challenge, it seemed to work. Investors realized that withdrawing their funds would immediately turn paper losses into actual (cash-flow) losses that would have to be reported to the public.

Conversely, the haircut Citron used to stem the run by OCIP investors could not be used with creditors, and creditors were champing at the bit to get back their funds as soon as possible. As interest payments on Orange County’s outstanding debts came due, most creditors refused to roll over their credits and others demanded back existing loans. Credit Suisse First Boston (CSFB), for instance, which had lent Orange County $2.6 billion in connection with reverse repo agreements, requested the repayment of $1.25 billion.133 Orange County was having difficulties meeting the rising demand for funds, and no one was willing to lend it such amounts. As a result, finding sources of liquidity to meet the investors’ growing demands for cash became of paramount importance.

Attempts to gain the needed funds by selling OCIP’s entire portfolio were unsuccessful, and appeals to the Securities and Exchange Commission (SEC), state treasurer’s office, California attorney general, and Governor Pete Wilson all were ignored. Everywhere Orange County looked, pockets were empty, and the predominantly Democratic legislature in California had relatively little sympathy for the rich, Republican residents of Orange County.

Government Paralysis

The government of Orange County needed to act promptly, but the Brown Act, which was passed in 1953 to protect the community’s right to attend and participate in public meetings, hindered swift action. The act required seventy-two-hour notification for regular meetings, twenty-four-hour notice for special meetings, and one-hour notice for emergency meetings, but financial crises did not qualify as “emergencies.” Criminal penalties were imposed on public officials who did not follow the rules. When news of the financial crisis first broke, Orange County’s Board of Supervisors should have moved quickly, but they were caught on the horns of a dilemma. Due to the Brown Act, they had to provide the public with notification of any meetings, but if meeting announcements were made, they might fuel public fears and exacerbate the outflow of funds from OCIP.

Citron Resigns

When CMRA finished the audit of OCIP, it estimated the OCIP portfolio had lost approximately $1.5 billion. These losses were reported to the press on December 1, and on December 4, Robert Citron resigned as Orange County’s treasurer–tax collector. News of the losses and Citron’s resignation cast a dark shadow over financial markets nationwide, and capital flight from the fund became even more problematic. Municipalities tried to liquidate their deposits; counterparties to reverse repo agreements demanded additional collateral to back their outstanding transactions, or they seized billions of dollars of collateral to sell at market prices; creditors refused to roll over their loans or called them early. State, county, and national bond prices fell (i.e., yields rose) as investors attached higher risk premiums to bond issues. Orange County housing prices fell, and residents braced themselves for the effects of emergency budget cuts, higher fees, and increased taxes.

Lack of Liquidity Leads to Bankruptcy

The run on OCIP was much the same as a depositors’ run on a bank, and in an effort to meet the surging demand for cash, Citron was forced to sell assets that were rapidly declining in value. OCIP was caught in a death spiral. As assets were liquidated, further losses were realized, stimulating even greater capital flight from the fund. With nowhere to turn, insufficient cash on hand, and no white knights in sight, Orange County filed for bankruptcy on December 6, 1994. Its filing cited the need for a “centralized forum within which to resolve any and all competing claims while assuring that nothing related to the problems of the fund interferes with the county’s ability to provide any and all essential services that the residents of Orange County have come to expect from their government.”134

Fire Sale of the OCIP Portfolio

Soon after bankruptcy was declared, Orange County appointed Tom Hayes, former California state treasurer, to reorganize the county’s finances. After receiving permission from the U.S. Bankruptcy Court, Hayes set his sights on selling the portfolio with as little negative market disruption as possible and investing the recovered funds in safe, short-term assets. Hayes hired Salomon Brothers to help, and, within six weeks, OCIP’s assets were sold. Ironically, the sale of OCIP’s assets received support from an unexpected source. In December 1994, the Mexican peso crisis led to the flight of international capital from developing nations to the safety of the U.S. dollar. The increase in the demand for U.S. financial assets raised security prices, adding to Orange County’s earnings.

Was Orange County Truly a Derivative-Related Failure?

Orange County’s troubles are often referred to as a derivatives disaster. Nevertheless, even though derivative-related securities accounted for a portion of the decline in OCIP’s portfolio value, Orange County’s misfortune appears to be related more to leverage, bad management, and relatively large holdings of fixed-income securities than it was to mistakes from investing in derivative instruments. Table 5.9 shows the composition of OCIP’s portfolio on December 1, 1994, the day it reported a $1.5 billion unrealized loss. OCIP held no forwards, futures, options, or swaps. The decline in its portfolio value was due to three main causes: structured notes ($722.8 million), fixed-income securities ($635.4 million), and reverse repo agreements ($67.1 million). Of these three sources, the losses on fixed-income securities and reverse repos were not derivative-related, but due to the rise in market interest rates and the reduction in spreads. Only if you considered reverse repo agreements to be a form of forward contract could you classify these losses as derivative-related, and even then, they would account for less than 5 percent of the total decline in OCIP’s portfolio value.

Table 5.9: Orange County Investment Pool Portfolio: December 1, 1994 (Millions of Dollars)

Source: Merton H. Miller and David J. Ross, “The Orange County Bankruptcy and Its Aftermath: Some New Evidence,” Journal of Derivatives 4(4) (Summer 1997), 53.

Of the losses on structured notes, inverse floaters were the major culprits, accounting for more than three-quarters of the $722.8 million decline in portfolio value, but were all of these losses truly derivatives related? A floating-rate note (with a bullet repayment) and an interest rate swap that has double the notional principal can replicate an inverse floater. Therefore, rising interest rates caused both the floating-rate portion and the derivative portion of the structured note to change. To calculate the pure derivative-related losses from these structured notes, you would have to separate the value of the derivative from the floating rate note. Such an analysis would reduce the size of the derivative-related losses somewhat below the figures given in Table 5.9. Regardless, the loss of anywhere near $722 million (for all structured notes) is substantial by virtually any comparative measure of derivative-related failure.

Another way to address whether Orange County’s losses were mainly derivative-related is to rely on information provided by Robert Citron prior to the failure. When he was interviewed, in April 1994, Citron revealed that only 20 percent of the OCIP portfolio was invested in derivative-related assets.135 If that figure were accurate and if these derivative positions resulted in losses proportional to their share of the portfolio, then, at most, $328 million of the total $1.64 billion of losses were derivatives-related. Losses of such magnitude would put Orange County into a special class of derivative-related failures, but it also would scale back dramatically the notoriety of the failure and weaken its relationship to derivative instruments.

Was Orange County Really Bankrupt?

The U.S. Bankruptcy Court ruled that Orange County’s declaration of bankruptcy was valid, and the county deserved Chapter 9 protection from creditors wanting immediate repayment, investors trying to cash out their investments, reverse repo counterparties seeking additional collateral to back their loans, and other parties with whom Orange County had a long list of financial obligations.136 Two financial measures that can give insight into the economic health of Orange County when it declared bankruptcy are solvency (i.e., if the county’s assets exceeded its liabilities) and liquidity (i.e., if the county had sufficient cash assets to meet demand).

Was Orange County Insolvent?

After careful analysis, the court decided that Orange County was insolvent. Nevertheless, despite its thorough analysis, lingering doubts remained as to whether this decision was correct. Table 5.9 (column 3) shows that on December 1, 1994, the market value of OCIP’s assets was approximately $6.1 billion greater than the value of its reverse repo agreements, indicating that Orange County seemed far from insolvent. With a positive net value so large, it is hard to understand the basis on which the U.S. Bankruptcy Court made its determination.137

Was Orange County Really Illiquid?

On December 1, 1994, the day its catastrophic losses were reported, OCIP had almost $650 million of cash (i.e., highly liquid assets) at its disposal (see Table 5.9). These funds could not have satisfied all the cash demands for withdrawals, early pension fund redemptions, calls by counterparties for increased collateral, threats from banks to withdraw credit lines, and the payment of operating expenses (e.g., salaries). Therefore, granting Orange County relief from its creditors was probably a reflection of justifiable liquidity needs. Nevertheless, OCIP had plenty of marketable securities it could have sold, along with approximately $32 million in monthly net interest earnings and its normal cash inflows of deposits from mandatory investors. As a result, the liquidity situation at OCIP may have been bad, but it may not have been as bad as it first appeared, which raises questions about the basis on which the U.S. Bankruptcy Court made its determination.138

Was It a Mistake to Liquidate the OCIP Portfolio?

Debate concerning the liquidation of OCIP’s portfolio is moot because, to a large extent, liquidation was not a choice Orange County made on its own. As the county defaulted on its debts, OCIP assets were confiscated by counterparties and sold at market prices. Of the total OCIP portfolio, more than half was liquidated in this manner.

Suppose investors and creditors were (somehow) convinced to wait. Figure 5.4 shows that the OCIP portfolio was liquidated at precisely the worst time possible. Interest rates in December 1994 peaked at just over 7 percent, and, during the next year, they declined by nearly 2 percent. After Orange County declared bankruptcy, OCIP’s entire investment strategy changed. Previously, Citron earned his profits by exploiting the upward-sloping yield curve (i.e., borrowing short term with reverse repo agreements and investing in higher-earning, medium-term assets), betting on declining interest rates, and leveraging the OCIP portfolio. The new strategy was to liquidate all existing positions, invest funds in money market assets, and unlever the OCIP portfolio.

If the administrators of OCIP had waited, slowly reduced the portfolio’s leverage, and taken measured steps to liquidate assets, results would have been remarkably different. By one estimate, if OCIP had pursued this hold-and-slowly-liquidate strategy from December 1, 1994 to March 29, 1996, the portfolio would have recovered approximately $1.8 billion of its value—an increase that would have eliminated all the losses reported on December 1, 1994 and contributed a significant positive return.139

The problem was that no one knew in December 1994 how future interest rates would change. Had they increased, OCIP stood to lose considerably more than it already had, and further losses could have impoverished Orange County’s already cash-strapped municipalities.

Could the Debacle Have Been Predicted?

Even though the future is unknown, reasonable estimates and efforts to moderate risk can still be made. For instance, Value at Risk (VaR) analysis might have been helpful to all the affected parties, by clarifying the risks facing OCIP. Based on OCIP’s December 1994 portfolio, VaR analysis would have shown that OCIP faced a 5 percent chance of losing at least $1.1 billion (in addition to the existing losses) during the coming year.140 The same figures would also have shown that OCIP faced a 5 percent chance of earning at least $1.3 billion.141 At a 5 percent level of confidence, a VaR of $1.1 billion means that once in twenty years there should be losses greater than or equal to $1.1 billion. Ironically, twenty years was about the tenure of Citron as Orange County’s treasurer–tax collector. Even if they had known OCIP’s VaR and internalized its 20-to-1 meaning, it is still uncertain whether Orange County’s Board of Supervisors and Administrative Officer would have meaningfully challenged the risks that Citron was taking.

As it turned out, interest rates fell precipitously during 1995, thereby catapulting OCIP into the upper tail of the distribution, but OCIP administrators had no way of knowing this in December 1994. Faced with an emergency, the possibility of losing significantly more, and only the remote prospect of turning its losses into gains, the administrators decided that the risk of waiting and liquidating the portfolio slowly was too high.

Robert Citron

In large part, Orange County’s bankruptcy was the result of a recklessly large bet by Robert Citron, but it was also caused by incompetence. After the dust settled, it became clear that Citron was an inept manager of the leveraged portfolio he created. Despite twenty-four years of experience, Citron had not acquired the skills or staff needed to price, track, or hedge the speculative positions he took. In effect, Citron ran an out-of-control hedge fund, but up until the end, he believed that it was the world, and not he, that had gone mad. He was convinced that Orange County had suffered only “paper losses” that never would have to be realized.

Court testimony added to the pile of evidence that Citron was unable to manage the OCIP portfolio. Psychologists and Citron’s attorney testified that Citron, the “financial wizard,” had only seventh-grade-level math ability and performed so poorly on tests that he “bordered on brain damaged.”143 Further testimony revealed that he suffered from a form of dementia and often relied on a mail-order astrologer and a psychic for his interest rate forecasts.144 Chillingly, Ernie Schneider, Orange County chief administrator, rationalized Citron’s behavior by saying, “Bob was always very unusual, eccentric, different, but I attributed it to, you know, brilliant psychics. Chess masters are not normal people.”145

On April 27, 1995, after a fifteen-minute arraignment, Citron pleaded guilty to six felony counts and three special enhancements.146 He was led away handcuffed, booked at the county jail, and then released on his own recognizance, with bail set at $25,000. Under state sentencing guidelines, Citron faced a maximum prison stay of fourteen years and a $10 million fine. The minimum penalty was six months in jail. Most of the charges against Citron related to his misallocation of returns, misappropriation of funds, and misleading/false statements. Citron was not charged with embezzlement, conspiracy, collusion, fraud, or trading for personal gain.

In the end, he was sentenced to serve one year in the county jail, pay a $100,000 fine for his financial shenanigans (e.g., siphoning off the interest earnings from investments made by schools, cities, and some small agencies and placing them in the county-run investment account), serve five years of supervised probation, and perform 1,000 hours of community service. Citron ended up serving just nine months in a work-release program, which means the seventy-one-year-old never spent a night in jail. Rather, he worked in the prison during the day but was allowed to go home in the evening. The light sentence was due largely to Citron’s open remorse throughout the trial, his cooperation with investigating authorities, and his deteriorating health.

In the trial, Los Angeles Superior Court Judge J. Stephen Czuleger reprimanded Citron for “gambling with public money” and chastised him for failing in his oath and obligations to the community. To Czuleger, the loss of public funds bothered him less than the damage Citron had done to the credibility of public officials.

Other Players: Matthew Raabe and Merrill Lynch

In his three-page plea, Citron implicated Matthew Raabe, former Orange County assistant treasurer, claiming that Raabe assisted in falsifying OCIP’s annual financial statements and lied about the fund’s financial condition. In 1996 (nineteen months after the bankruptcy), Citron went further and testified that Raabe was responsible for conjuring up the scheme to divert funds from cities and special districts to Orange County’s budget.

On May 5, 1997, Raabe was convicted on five counts of fraud and misappropriation of $88.5 million and sentenced to three years in the state prison. After serving only forty-one days in jail, he was released, and in November 1997, his conviction was overturned on grounds that the former district attorney (Mike Capizzi) should not have prosecuted the case, due to his heavy involvement with the Orange County bankruptcy. In 2001, the California Supreme Court refused to review the overturned conviction. Raabe was free.147

Orange County filed a civil suit against Merrill Lynch, its major broker/dealer, on grounds that it sold improper and illegal securities to Citron and made false statements in its securities sales. Orange County prosecutors even recruited Citron to help develop their case. Merrill Lynch claimed that its conduct with Orange County was proper and professional and accused Orange County of “letting the real culprit off because he’s going to help you go after the deep pockets.”148 In 1998, the case was settled out of court.

Stealth Supervision: Shared Blame

Even though Robert Citron was the chief protagonist in the Orange County bankruptcy, the lack of adequate supervision and control of the risks in OCIP’s portfolio were significant contributing factors. Orange County’s Board of Supervisors, the SEC, credit rating agencies, media, and residents of Orange County were all looking in the wrong direction when the fatal blow was struck.

Many people wondered why the SEC, which met with Citron in 1994 and conducted multiple independent audits throughout 1994, found nothing unusual in the OCIP portfolio. They were also baffled that the two largest U.S. credit rating agencies, Standard and Poor’s and Moody’s Investor Service, gave Orange County’s portfolio, in 1994, their next-to-the-highest ratings (AA and Aa1, respectively). The reason for Orange County’s clean bill of health appears to be, in large part, because at each of the levels of overview, focus was put mainly on Citron’s return and credit risk, rather than on the risks that brought on the Orange County collapse (i.e., market risk and liquidity risk).

Especially disappointing was the oversight role played by Orange County’s Board of Supervisors. During the court proceedings, one Orange County supervisor admitted that his “knowledge of public finance matters was nil prior to the bankruptcy”; he never read any of Citron’s reports and resolutions requesting approval for hundreds of millions of dollars in bond issues, and he rarely asked questions because “I wouldn’t know the kinds of questions to ask.”149 In December 1995, the Orange County grand jury charged Board of Supervisors members Roger Stanton and William Steiner and auditor-controller Steven E. Lewis with willful misconduct, but these charges eventually were dismissed by the courts.

Governance Reforms

Having been burnt once and wanting to ensure it did not happen again, Orange County implemented a series of financial governance reforms. The bankruptcy spurred the establishment of a crisis management team, and the county appointed a chief executive officer and chief financial officer to make serious budget cuts and bring its financial house in order.

A package of state and local reforms was passed that mandated more transparent and timely reporting and oversight. An internal auditor was appointed, reporting directly to the Board of Supervisors. Public officials were required to make open disclosure of possible conflicts of interest. Prohibitions were imposed on oversight committee members from receiving gifts. Austere investment guidelines were introduced, which put top priority on acquiring assets with relatively low risk and high liquidity. Portfolio managers were constrained from leveraging investments (e.g., entering into reverse repo agreements), transacting derivative deals, and purchasing structured notes. The selection of brokers, dealers, and investment advisors also received greater scrutiny.

Safety, Liquidity, and High Yield Are an Impossible Combination

One of the most basic relationships in finance is the direct relationship between risk and expected return. In markets where billions are invested, it is difficult to earn returns that consistently beat the market average, and investment funds that do are most likely carrying greater than average risks. For this reason, risk and return should be reported simultaneously to investors for rational choices to be made. To this end, portfolio managers should be required to communicate (and investors should demand) timely, transparent information on the risk and return profiles of their portfolios.

Value at Risk (VaR) analysis could have aided greatly in this effort because it communicates to investors the odds of losing a certain amount (or more) during a given period. So long as the investors understand these risks and are willing to bear their downside consequences, mistakes as well as the grounds for legal suits and crushed expectations are diminished.

If You Can’t Explain It, Then Don’t Do It

As financial instruments become more complex, valuing them and tracking their risks requires financial sophistication that is beyond the abilities and/or time constraints of most investors. Even though investors in OCIP and regulators can be criticized for not demanding better explanations of Robert Citron’s investment strategies, Citron is culpable for his inability to articulate them clearly. As intricate as investment strategies might be, there should always be a way of boiling them down to a short, informative overview. For instance, Citron might have explained his strategy like this:

In the past, OCIP beat the average market return because medium-term interest rates remained above the short-term rates and because interest rates fell. During the coming year, we believe these conditions will continue, and therefore the OCIP portfolio is structured to profit from these two expectations. To capitalize on our outlook, we have leveraged the OCIP portfolio by borrowing about $2 for every $1 you have invested in the pool. More specifically, OCIP now has about $20.5 billion in total assets, of which $13 billion were purchased with borrowed funds, and the rest was financed with your (investor) funds. This (nearly) 3:1 leverage factor increases substantially the amount of risk you are facing.

As far as risks are concerned, OCIP’s assets have virtually no default risk because the portfolio contains only the best creditors (like the U.S. government), but OCIP does face a significant degree of market risk and liquidity risk. Market risk is the potential for OCIP’s portfolio value to fall due to rising interest rates. Liquidity risk is the potential that we will not be able to meet the cash demands of investors, creditors, and/or investment counterparties. By our best estimates, there is a 5% chance that the currently structured portfolio could lose $1.1 billion or more during the next year and an equal chance that it could gain $1.3 billion or more during the same period. Seen another way, as of November 1994, the OCIP portfolio stands to win or lose about $550 million for every 1% change in market interest rates.