Revenue Rise with Loans

At this point a vice president at the mortgage servicer urged me to work on improving their process. The first thing I did was bring the servicer and custodian together to talk about the tracking issue. After several meetings, we agreed to try working together to integrate their two processes. Using the Partnership Continuum model, we started off by assessing and defining our needs. Once both parties agreed that they needed to get the correct document delivered to the correct location in time for the closing, we created a vision for the partnership. We then started on the initial activity: to document the two organizations’ processes, set standards, and create a measurement system.

After several months, the firm of mortgage servicers was able to improve its on-time delivery to closers by 200 percent. After one year, it had one of the best records in the business. Bob later confided that initially he hadn’t believed we’d improve things.He’d felt that the custodians had spent too much money on their delivery system without
consulting customers and that they would never change it. But partnering with the custodian and creating a sense of interdependence proved otherwise.

The asymmetric distribution of corporate bond returns is easily explained by the Merton model introduced earlier. Numerous studies substantiate that even the return distributions of less risky and more liquid asset classes like government bonds are skewed and leptokurtic. Moreover, index returns exhibit significant autocorrelation that can be explained partially by a permanent component. Basically, bond returns are a result of price movements, interest accrual, pull to par, and roll down of the yieldcurve effects. While the first component is highly variable the other three
components are rather stable over time.

Because of their long history and good data reliability the empirical study is based on Merrill Lynch indices for the period January 1987 to September 2003. So the sample period comprises 201 months, spanning more than two business cycles. It contains the 1989/90 US recession, two periods of dramatic Fed tightening, the Tequila crisis in 1994, the Asian crisis in 1997, and Russia’s default in 1998. Driven by a secular trend of disinflation the yield of 10-year treasury notes declined from 7.2 to 3.9 percent in this period. With regard to future return expectations this should be kept in mind.

A special situation occurs in the high-yield sector. The illiquidity in large parts of the universe causes price lags meaning that the aforementioned small changes in credit quality are not immediately reflected in bond prices. In the economic literature, this effect is known as non-trading. With respect to high-yield indices non-trading and non-synchronous trading of index bonds can lead to autocorrelation of returns. Therefore, estimates of the index volatility for illiquid asset classes can be distorted. Usually the true volatility of illiquid asset classes is underestimated. In a portfolio context too large portfolio weights are the consequence. Especially portfolios constructed in the classical mean–variance framework suffer from that problem.

Mean–variance analysis, made popular by Markowitz and Sharpe, has been the basis for the process of portfolio optimization since the 1990s. Yet, the method itself suffers from various pitfalls. Among others it ignores deviations of the return distributions from normality. The asymmetric risk profile of corporate bonds and the illiquidity of certain segments of the international corporate bond markets make great demand on the process of portfolio construction. Merton (1974) clarified that corporate bonds can be replicated by the combination of a riskless bond and a short put on the assets of the company. This shows that the return potential of corporate bonds is somewhat constrained whereas the possible loss in the event of default is only limited by the recovery. Between 2000 and 2002, spectacular defaults like Enron and WorldCom heightened the sensitivity of investors to the risks associated with credits. In general the following relation holds: the higher the leverage of an issuer, the higher the credit risk. Remember that the short-put option on the assets of a highly leveraged issuer is much closer at-the-money than that of a conservatively financed company. And the closer the short put option is at-the-money the more asymmetric becomes the risk profile of a corporate bond. For those issuers small changes in credit quality such as, for example, due to increased dividend payments, can lead to significant volatility of spreads and corporate bond prices.

Clearly investors willing to allocate a part of their budget to corporate bonds are facing two questions. First, they have to decide how much of their budget they want to invest in corporate bonds. In this context we will focus on a pure fixed income portfolio. Usually private as well as institutional investors define their long-term asset allocation with respect to the asset classes such as stocks, bonds and real estate in a preliminary process.

The second decision concerns the sector allocation of the portfolio which is part of the tactical asset allocation. As noted before, the sector allocation of a corporate bond portfolio essentially determines its risk/return profile and portfolio beta.

Having focused on indicators for the overall level of leverage so far, we will now switch to metrics that relate the ability to generate cash flows and profits to the interest burden. This helps to better capture liquidity problems in the short term, but goes at the expense of understanding the longer term vulnerability of the corporate sector due to leverage. Although facts seems to show a long-term upward trend in the ratio of net interest payments to cash flows, one can argue that the level reached in 2003 roughly represents the average over the last 35 years. The rise in the overall level of indebtedness in this period has coincided with significantly falling interest rates, keeping the interest burden for the companies on a manageable level.

Like the ratio of net interest payments to cash flows, interest coverage, defined as the ratio of profits to net interest payments, has remained in a range since 1970. Again, as one might expect, when net interest payments are better covered by the level of profits, creditworthiness improves, and credit spreads tighten. But interest coverage in 2003 was still near its cyclical low, indicating weak profitability in the nonfinancial corporate sector and a lack of top-line growth. Considering the fall in yields, the low level of interest coverage is problematic. Partially it results from the fact that not only the companies’ liabilities are interest-rate sensitive, but also the value of a part of the assets depends on the level of interest rates.

While the choice of the most appropriate measure of corporate leverage is an arbitrary task, empirical studies indicate that the financing gap is able  to explain a lot of the variance in credit spreads. It is defined as the difference between capital expenditures, including outlays for inventories, and the amount of cash that corporations need to raise in order to finance their investment plans, expressed as a percentage of nominal GDP. The rapid rise in this metric was a powerful warning sign at the end of the 1990s that capital spending was overextended. As one usually observes at the end of recessions, the financing gap has not only closed after the 2001 recession, but fell substantially below its long-run average. In 2003, after-tax cash flow exceeded capital outlays, which is a rare occurrence. In the past, this was usually a good sign for capital spending. Intuitively, the US Treasury replaced the corporate sector as a financier for business spending.

Via tax cuts it provided the corporate sector with capital, thereby raising after-tax cash flow, and reducing the requirement for external financing through corporate debt issuance. The increasing amount of Treasury supply, combined with deleveraging in the corporate sector contributed significantly to corporate spread tightening between autumn 2002 and 2004.

Yet, free cash flow numbers are not available for all companies. European firms, for example, historically have published only profit figures. Therefore, we also look at EBIT and EBITDA figures. Of these two the latter is preferred because it deducts cash that is needed to maintain the operations. Specifically, on the macro level we use pre-tax profits with inventory adjustment and capital consumption adjustment, and add net interest payments to end up with an aggregate measure of EBITDA. The ratio of debt to EBITDAbehaves similar to the debt to cash flow metric, albeit it is considerably more volatile. Part of this volatility is induced by the degrees of freedom related to the accounting of depreciations and amortizations.

To forecast profits, one should look at margins and the macroeconomic prospects for growth. Popular proxies for profit margins are the difference between annual percentage changes in the GDP deflator and unit labor costs, and Merrill Lynch’s corporate misery index.

So far, we have identified a close correlation between credit spreads and future economic activity. When the economy is growing, usually each sector benefits from that growth. In the labor sector, economic growth typically leads to falling unemployment and rising wages. Historically, the corporate sector has benefited particularly during economic expansions. Not only did corporate profits grow in those periods, but also they increased faster than national income. This is reflected by a rising share of national income that usually goes to the bottom-line of the corporate sector during economic expansions. In other words, corporate profitability rises. In recessions, conversely, the labor sector as well as the corporate sector suffers. The problems of the corporate sector become evident by a declining ratio of profits to GDP. Hence, corporate profitability and thus the companies’ abilities to generate cash flows to service their liabilities are highly cyclical. The cyclical nature of corporate profits and cash flows explains variations in corporate bond spreads extremely well. In our view the most objective measure for the ability of companies to service their liabilities is free cash flow. As a proxy for the overall free cash flow that is generated in the US corporate sector, the flow of funds statistics provides a measure that is known as internal funds.

Generally, in growing economies the nominal level of debt rises over the years. This is mostly because of inflation and a steady growth in balance sheets. The long-term upward trend has accelerated sharply since 1980, coinciding with the sustained fall in short- and long-term interest rates and inflation rates. Since then, the steadily falling level of interest rates has allowed companies to take on significantly higher levels of debt than before. Despite record debt levels, interest payments have remained manageable.

To gain a better insight in the debt burden, we have graphed the level of credit market instruments outstanding relative to the GDP of the nonfinancial corporate sector. The second line shows the level of indebtedness in real terms, that is, after adjustment for the price deflator of the nonfinancial corporate sector. Both measures exhibit a cyclical pattern that is overlaid by a long-term uptrend. Usually indebtedness rises in the first stage of a recession and is cut back consequently at the end of each recession and the following years, reflecting the companies’ efforts to restructure their businesses and clean up their balance sheets. In the later years of an expansion top-line growth usually slows and companies tend to increase their leverage to boost return on equity and to finance future growth, for example, through acquisitions.