A Slightly Different Way to Prevent Bank Runs

A few weeks ago, while reading yet another book on the financial crisis, I had the beginnings of an idea about how to help financial institutions who have experienced a decline in assets relative to their liabilities. I thought about possible ways to recapitalize these firms without direct government intervention, as happened during the TARP. The idea I came up with was a quasi-convertible bond to be forcibly converted by a regulator during a time of crisis.

This bond is one that can be issued by a financial firm, usually an investment bank, in order to raise funds to use for trading and other proprietary investments. The bonds will have an intermediate to long-term maturity ranging from five to fifteen years. Under normal circumstances the bonds will act in a similar fashion to current debt issues. However, the bonds will have a provision whereby they automatically convert to equity when their government regulator determines that the firm is under capitalized or facing a liquidity short fall.

The purpose of this conversion is to quickly, and forcibly, lower a firm’s liabilities and increase its capital cushion. Such an increase in capital would allow the firm to maintain a sufficient liquidity position and prevent a need for capital injections or support by the government should the firm become sufficiently illiquid. This type of bond would allow financial companies to finance their normal functions under normal credit conditions, but allow the firm to recapitalize itself under excess financial strain.

The decision to convert the bond into equity would rest with the firm’s main regulator, the Treasury and/or the Federal Reserve depending on the type of firm. For example, the Federal Reserve Bank would decide institutions that maintain customer demand deposits. The government decision makers would base their decision on the firm’s capital strength, counterparty risk and balance sheet stability. If the firm is deemed to be illiquid, the debt will be forcibly converted to equity and creditors would become common shareholders. Several tranches of these bonds could be sold for any individual firms. The highest risk could be converted firms, followed by subsequent levels. The government regulators would determine the necessary ratio of this type of debt to overall debt issued by each firm. Likewise, during a conversion, the regulator would also determine the amount of the bonds to be converted.

The purpose of these bonds would be to protect the government from having to use taxpayer funds to bailout or otherwise support solvent but illiquid firms. If a firm begins to run out of liquid securities to post as collateral, it has a liquidity problem. During a crisis, firms become insolvent as they are forced to sell its assets to acquire necessary liquidity, often at a significant loss. A forcibly convertible bond would allow for the firm to tap a reserve of liquidity that would be sufficient for it to ride out the crisis. If the firm is not able to acquire the necessary liquidity, the government can be forced to put taxpayer funds at risk to protect the firm’s counter-parties from its collapse.

Obviously, this type of bond will be less desirable to debt purchasers as it would essentially force them to assume greater risk associated with a firm’s activities. To make up for this risk, the bonds would likely trade at a discount to normal debt. As a result, the government would have to use its regulatory powers and force firms to issue this type of debt. This would prevent the government from having to use taxpayer funds in the future to backstop an illiquid bank during a financial crisis by recapitalizing that bank when it is short of necessary liquidity.

Game Theory and Census Workers

I'd like to make a short thought exercise which might be an example of economic problems which the US census might ultimately cause. And, no, I am not trying to say that the census will cause problems but there might be unintended consequences.

Consider you are either an individual who has been unemployed for a long period of time or a new graduate just trying to enter the job-market. You have had a very difficult time finding stable work, but have found an opportunity to work for several months. This job offers you the chance to work from twenty to forty hours per week, and the choice is entirely yours. Along with hundreds of thousands of others in your situation, you've decided to take this job opportunity.

Being a person who is in need of money who would like a full time job, instead of working about twenty hours a week you would likely decide to work towards the forty hour mark. You do your work eagerly and try to get as many hours as possible to maximize your income. Much like you, the other hundreds of thousands of people likely decide to follow a similar course and work as many hours as possible.

So, here we are. Individuals who are working for the census have every incentive to maximize their hours worked. Unfortunately, there is a limit to the amount of work these people can do since most census workers were hired to go door-to-door and count those who have not filled out their forms. There were only so many people who failed to return their documents, so there is a upper limit to the trips census workers must take. With employees attempting to work as many hours each week as they are allowed, they will quickly visit all of those who's forms are outstanding.

The unfortunate consequence to this exercise is that census workers will likely complete their door-to-door visits well ahead of schedule. As the documents are gathered in each area, the workers will be released from their employment earlier than they were expecting. The census has caused a major boost in hiring and has helped to lower the national unemployment rate. As the newly hired census workers finish their jobs, they will be released from their employment and will reverse any positive effects from their hiring. The unemployment rate will likely again spike as many hundreds of thousands of former census workers file new claims.

In the coming months, the unemployment rate will increase even though the economy is improving. Potentially, as these people again lose their jobs there can be severe negative effects on the national economy. Companies get scared when unemployment rises and consumers spend less when they lose their jobs. The ultimate loss of employment may cause these concerns to return and output to decrease. The census jobs, once lost, can potentially damage the fledgling economic recovery for these reasons.