New View of Citi
Previously, I suggested that one of Congress’s central goals for adopting financial market reform is to limit systemically significant financial institutions’ risk taking to levels where the capital and assets of each business covers its losses. In other words, the reform bill aims to require businesses to internalize both their successes and the consequences of their failures. Some question the federal government’s expertise in risk management matters, while others challenge the appropriateness of government intervention into a corporation’s internal affairs. These challenges raise important questions. Before reaching these questions, however, it may be useful to consider a recent example of failed risk management that has prompted government intervention and the proposed reform bills’ potential impact on enterprise risk management.
In November of 2008, the federal government agreed to invest $45 billion (later converted into $25 billion of Citigroup common stock) and agreed to guarantee $306 billion in Citigroup’s loans and securities. (See here and here.) After suffering devastating losses beginning in the final three months of 2007 and continuing through the end of 2008, the esteemed investment bank with a 100-plus year history teetered on the brink of collapse. How had Citi’s star fallen so far so fast?
Citigroup had one of the largest subprime mortgage portfolios held by a Wall Street investment bank – at one point the portfolio totaled more than $58 billion. Citigroup also held a significant amount of securities related to residential mortgages and acted as one of the largest distributors of collateralized debt obligation (CDO) securities –investment products that combined groups of residential mortgages and sold the rights to the cash flows from the bundled products as exchangeable securities. The owners of the CDO securities received access to the money collected in connection with mortgage payments. Hence, when subprime mortgage holders began to default on principal and interest payments on their mortgages, Citigroup contemporaneously suffered significant losses in several of its business units.
Why wasn’t Citigroup able to navigate through the dangerous swamp of exotic financial products and excess leverage? Citigroup had employees assessing the risks related to each of its investments and a team of senior managers, executives (including one of the most highly compensated CEOs in the industry) (see here and here) and a prominent board of directors assessing its enterprise risk levels.
Citigroup’s near collapse and its subsequent federal bail-out illustrate elements of both cognitive biases and limitations in the accuracy of risk models. For several months after other companies reported losses related to the subprime mortgage markets and investment strategies linked to subprime mortgage markets, Citi executives ignored the warning signs and continued with its investment strategies. Lucrative compensation and promising projections likely inspired some Citi executives’ to take a short term view on the appropriate levels of enterprise risk. A few executives have argued that they made poor investment decisions justifiably relying on credit rating agencies’ assessment of the value of CDO securities. Still others assert that they relied on quantitative models that (based on an inaccurate assessment of the possibility of a national decline in U.S. home values) wrongly suggested little or no probability of loss on these residential mortgage related securities.
Would any of the provisions of the currently proposed legislation have helped Citigroup better manage its enterprise risks? Proponents of the legislation argue that a provision establishing a collaborative oversight council of federal regulators could have helped to identify the danger in Citigroup’s residential mortgage investment strategy and its exposure to higher risk investment products sooner and prevented some of its losses. It is possible that a group of regulators could have accurately assessed the failures in Citigroup’s investment strategies (other investment banks and hedge funds did figure out that the market was headed downhill and they changed their exposure to bet against the market). It is, however, difficult to imagine the mechanics of regulatory oversight functioning in this manner or to envision how such oversight might influence independently developed and rapidly changing investment strategies. The mechanics of such regulation are unprecedented, but not impossible. Stricter standards imposed to ensure safety and soundness may have at least curbed the firm’s involvement in the CDO market. The impact of the regulatory council would also depend on a high level of cooperation and collaboration among federal agencies. Greater interagency cooperation is a valuable and feasible goal.
Presumably, the Volcker Rule would have expressly prevented Citigroup from engaging in the higher leverage, riskier, more exotic investments products. The provisions of the proposed legislation demanding greater transparency in the sales of instruments like credit default swaps, an insurance-like product that exacerbated Citigroup’s exposure to the residential mortgage market, may have help Citigroup’s investment professionals, its board and its shareholders better understand the risk levels associated with these investments. Finally, a provision in both the House and Senate versions of the financial reform bill would have provided for orderly liquidation of Citigroup “so that creditors and shareholders bear the loss[es]” and no future federal fund bail-outs would be necessary. The effectiveness of this last-will-and-testament provision is hotly debated.
While the nuances of risk management make it difficult to conclude that the provisions in the financial reform bills will be sufficient to prevent future crises, the hypothetical application of the provisions to Citigroup’s decisions prior to and during the crisis suggest that a few of the measures may have been useful to shore up Citigroup’s enterprise risk management policies. Until the bill has been finalized, however, assessing even the possibility of its effectiveness on enterprise risk management issues may be premature.