What is the right amount of debt for a particular borrower? This depends on the borrower.  Are we talking about a corporation, a bank, a household, a government entity, or an investment fund?  At a very high level, it seems that some debt can be beneficial, while too much debt can be detrimental.  Sort of like the story of the three bears.  Let’s start with a non-financial company.

What is the correct amount of debt for a non-financial company to carry? In corporate finance theory, the mix of debt and equity on a company’s balance sheet is called its “capital structure” and the ratio of debt to equity is called “leverage.”  The optimal leverage is that which maximizes the value of the company.  In the 1950’s, Nobel Prize winners Franco Modigliani and Merton Miller showed that in a special world with no taxes and no bankruptcy costs, the value of a company is completely independent of its level of debt or leverage.  Capital structure does not matter.  However, in the real world, corporate taxes do exist and are applied to profits after interest expense is deducted.  This deductibility of interest renders debt attractive relative to equities, so companies are incented to issue more debt.  As leverage increases, the probability of financial distress or default increases as well.  This puts an upper limit on the feasible level of leverage.

The appropriate level of leverage for a particular company depends on the stability of its cash flows. Companies with unstable or unpredictable cash flows, like start-ups, will carry little or no debt.  Companies with stable cash flows, like utilities, utilize debt more heavily with leverage of two or three (that is, debt equal to two or three times the value of equity).  The greatest leverage is typically found in financial companies like commercial banks or investment banks.  The differences are remarkable.  The overall level of leverage for the universe of non-financial companies is 50%.  The average level of leverage for the banking sector is more than 20 times larger.

The proper amount of debt or leverage for a non-financial company is a decision by management, which is subject to lender willingness to lend, credit rating agency demands, and the level of management prudence. If a company becomes over-levered and undergoes financial distress, the solution is simply that the company is restructured with the equity holders largely or completely wiped out and the debtholders becoming the new owners.  There does not appear to be a compelling reason for regulators to get involved with the leverage decision for a non-financial company.

Things are different for financial companies. An important debate today revolves around the appropriate leverage for banks, particularly for large banks.   The specific issue is what should be the minimum level of “capital” (in bank jargon “capital” basically means common equity but can also include other accounts like loss reserves, preferred stock or certain types of bonds) relative to assets.  Many academics and regulators argue that banks should be required to carry much higher capital levels than they do today, while most bankers argue that greater capital requirements will reduce the return on equity (ROE) in banking and drive investors away from the banking sector.

An important lesson from Modigliani and Miller (MM) is that the ROE is not the best measure of a company’s financial performance. The ROE will generally rise with leverage; at least it will so long as the return on assets is greater than the cost of debt.  But what MM pointed out is that the cost of equity also rises with leverage, as the company’s stock becomes riskier.  This means that the ROE demanded by investors rises as well.  This argument is part of the academic response to banker concerns.

It is not only the level of debt that matters; terms of the debt matter as well. A major problem during the financial crisis was that many financial companies were funded with high levels of short-term debt that had to be rolled over frequently.  From a lender’s perspective, short-term debt is less risky inasmuch as it enables the lender to withdraw funds at the first sign of financial distress.  Prior to the crisis, much of this short-term debt was funding portfolios consisting of complex mortgage-backed securities (MBS).  Once concerns were raised regarding the true value of the MBS, lenders demanded greater collateral for existing loans.  This created a liquidity crisis for many financial institutions.  Short-term debt is less risky for the lender, but more risky for the borrower, particularly if it is used to fund long-term or illiquid assets.

For more than 30 years, there has been a concerted effort to come up with international guidelines for bank capital levels. This effort has been concentrated at the Bank of International Settlements in Basle.  The so-called Basle Accords (there have been several, first Basle I and then followed by, not surprisingly, Basle II and Basle III) are intended to set capital standards for banks that operate globally.  This enormous effort has been somewhat undermined by the financial collapse in 2007-2009, given that major banks required bailouts, even though they were compliant with the Basle guidelines.  The financial re-regulation bill that was passed following the crisis, the Dodd-Frank Act (DFA), imposed further capital and other restrictions on large U.S. banks.  One of these was a requirement that large banks undergo an annual “stress test” that is designed to reveal if the banks have sufficient capital to weather a severe economic scenario.  If a bank does not pass the test, it is required to increase capital ratios by issuing new equity, reducing dividends, reducing growth plans or other measures.

It is generally agreed that regulators should get involved with bank capital levels while it is okay to let the management teams at non-financial companies make capital structure decisions on their own. The reason is that bank failures (at least large bank failures) are believed to be much more detrimental to the overall economy than failures of non-financial companies.  The question then comes back to how much capital should we require the banks to hold.  I think the answer should be:  enough so that the probability of failure is negligible.  This will entail higher capital requirements than exist today, maybe much higher.  Then the next question will be:  do higher bank capital requirements impair economic growth?  This is a difficult question, involving the intricate relationship between economic growth and debt.  We’ll investigate these issues in future blogs.