So, if you are an entrepreneur who wants to start a company, what can you do to reduce the expected cost of capital? The answer is that it is in your interest to disclose to the lender all the information you can, provided you are the type of entrepreneur who is likely to pay back the loan. You want to reduce the lender’s doubt about future repayment. Unfortunately, this can be very difficult. The lender can neither peer into your brain, nor give you a good lie detector test. Attempts to convey information credibly in the real world are many, but there will always be residual information differences—they are just a fact of life. Still, if you can reduce the information differences, your firm will be able to enjoy lower costs of capital. Also, if you as borrower fail to give your best try to convince the lender of your quality, then the lender should not only assume that you are an average company, but instead assume you are the very worst—or else you would have tried to communicate as much as possible.
There are at least three important mechanisms that have evolved to alleviate such information differences. The first mechanism is covenants, which specify upfront what a debtor must do to maintain credit. This can include such requirements as the maintenance of insurance or a minimum corporate value. The second mechanism is collateral—something that the creditor can repossess if payments are not made. But anything that inflicts pain on the debtor will do. For example, if defaulting debtors were thrown into debtor’s prison (as they often were until the 19th century), the promise to repay would be more credible and lenders would be more inclined to provide funding at lower rates. Of course, for the unlucky few who just happened to suffer incredibly bad luck ex-post, debtors’ prison has some definite drawbacks.
The third mechanism to alleviate repayment uncertainty are credit rating agencies, which keep a history of past payments to help assess the probability of future default. This is why you need to give your social security number if you want to take out a substantial personal loan—the lender will check up on you. The same is true for large corporations. It may be easier to judge corporate default risk for large companies than personal default risk, but it is still not easy and it costs both time and money. The two biggest bond credit rating agencies for corporations are Moody’s and Standard&Poors. (The other two are Duff and Phelps and Fitch.) For a fee that the corporate borrower pays, they will rate the bond’s quality, which reflects the issuer’s probability that the bonds will default. This fee depends on a number of factors, such as the identity of the issuer, the desired detail in the agencies’ investigations and descriptions, and the features of the bond (e.g., a bond that will pay off within one year is usually less likely to default before maturity than a bond that will pay off in thirty years; thus, the former is easier to grade). The credit rating agencies ultimately do not provide a whole set of default probabilities (e.g., 1% chance of 100% loss, 1.2% chance of 99% loss, etc.), but just an overall rating grade. It is up to the ratings’ reader to translate the rating into an appropriate compensation for default risk. The top rating grades are called investment grade, while the bottom grade are called speculative grade (or junk).
There is often a sharp difference in quoted interest rates between the worst investment grade bond and the best speculative grade bond, partly also because many investing institutions are allowed to hold only investment grade bonds.
So, is there a difference between bonds of different rating quality? Yes! Altman studied corporate bonds from 1971 to 2003 and reported default and recovery rates. Very few investment grade bonds default—and especially right after issue when they would have still carried the original credit rating. However, many speculative bonds will eventually miss at least one coupon payment (which is considered default). Upon default, an AAA or AA bond price was worth about 75 cents on the dollar; an A bond price was worth about 50 cents on the dollar; and lower rated bonds were worth about 30 cents on the dollar.
Unfortunately, although bond rating agencies will update their rating if the condition of the firm changes, the empirical evidence suggests that these bond ratings are not very good in helping an investor earn superior rates of returns. In fact, the ratings seem to respond more to drops in the value of the underlying bonds than vice-versa. The bond rating agencies seem to be more reactive than proactive.
How do bond ratings translate into differences in promised (quoted) bond yields? lists the borrowing rates of various issuers in May 2002. (Many other current interest rates can be found at www.bloomberg.com/markets/rates/index.html and bonds.yahoo.com/rates.html.) Most of the differences between these borrowers’ promised interest rates and Treasury interest rates are due to default risk, which compensates lenders for differential default probabilities.
Where do such market imperfections apply? In the world of large corporations, the interest rate spread between similarly risky borrowing and lending rates is often mild, so they can pretend they live in a “perfect” market in which they can separate the project choice from their financial situation. Their promised borrowing interest rates would still be higher than what they can receive investing their money in Treasury bonds—but, given that these large firms still have some possibility of going bankrupt, their expected borrowing cost of capital would probably be fairly similar to the expected rate of return that they could earn if they invested money into bonds with characteristics similar to those that they themselves issued.
In the world of individuals, entrepreneurs, and small companies, however, expected borrowing interest rates are often higher than expected saving interest rates. In fact, this issue of an extraordinarily high differential between expected borrowing and lending rates—and with it the role of cash-on-hand—is one important difference between “ordinary corporate finance” and “entrepreneurial finance.” Entrepreneurs find it very difficult to convey credibly their intent and ability to pay back their loans. As a consequence, many entrepreneurs even resort to financing projects with credit cards, which may charge a thousand basis points above Treasury bonds. These high borrowing costs can thus prevent rational entrepreneurs from taking many projects that they would undertake if they had the money on hand. It also means that more established firms or richer entrepreneurs should optimally take more projects than poorer entrepreneurs.
But be careful in the real world before you conclude this to be the case: Entrepreneurs tend to have notoriously over-optimistic views of their prospects. (Even venture capitalists, the financing vehicle for many high-tech entrepreneurial ventures, which advertise returns of 30%/year or more seem to have managed to return only a couple of percentage points above the riskfree rate over the last thirty years.) This may actually mean that entrepreneurs face only high promised borrowing costs, not high expected borrowing costs. Thus, the quoted spread between their borrowing and lending rates, which is really all that you can easily observe, likelyhas a large component that is due not to information disagreements but simply due to credit risk.
The most obvious cause for different borrowing and lending rates are differences in opinion between the lender and borrower. To think about this particular assumption violation, we must work in a world of uncertainty—it would be absurd to believe that such differences in opinion could exist if there is no uncertainty. So, what happens if the lender and borrower have different information or different judgment about the same information? Most prominently, they could disagree about the default risk! For example, if you have no credit history, then a lender who does not know you might be especially afraid of not receiving promised repayments from you—from the perspective of such a lender, you would be extremely high-risk. Your lender might estimate your appropriate default probability to be 20% and thus may demand an appropriate default premium from you of, say, 8% above the risk-free yield. On the other hand, you may know that you will indeed return the lender’s money, because you know you will work hard and that you will have the money for sure. In your opinion, a fair and appropriate default premium should therefore be only 0-1%.
When your potential lenders and you have different opinions, you then face different expected savings interest rates and borrowing costs of capital. That is, if you know that you are a low risk, then your borrowing cost of capital (the expected interest rate) would not only be much higher than the lender perceives it to be, but it would also be the case that your borrowing expected cost of capital is much higher than your savings expected rate of return that you would earn if you deposited your money in the bank. You might be able to borrow at a cost of capital (expected rate of return) of 12%, but save only at an expected rate of return of 5%.
It is not just that you must offer a higher promised rate of return than what you expect to repay. Instead, you must now offer a higher expected rate of return when you want to borrow, compared with the expected rate of return that you could earn if you deposited money in the bank. (The bank is unlikely to go bankrupt, and your deposit is probably insured by the government, which means that the bank may pay not more interest than the equivalent short-term Treasury. The bank’s promised rate of return is almost the same as the expected rate of return.) The difference in stated borrowing interest rates and stated saving interest rates at your local bank could just as well be the default premium—the difference in the returns you promise and which you expect to pay, although in a different guise. The bank quoting you a lower savings deposit interest rate than borrowing loan interest rate would just compare promised interest rates, not expected interest rates. Instead, the novel issue discussed in our blog is that disagreements and information differences are now causing differences in expected returns. The borrowing and lending expected rates of return are no longer the same.Perfect Market Default Spread Assume you have a 5% objective probability of total default. In this case, a risk-neutral bank would ask you for a return of 5% · 0 + 95% · x = (1 + 10%) · $1, 000 ⇒ x ≈ $1, 158 . (6.1)
This 15.8% interest rate is not a market imperfection. It is merely compensation for your probability of defaulting. Both you and and the bank would realize that you are paying a 10% cost of capital.
Imperfect Market Credit Spread Now assume that you believe you have a 3% objective probability of total default, and the bank believes you have a 9% objective probability of default. In this case, the bank would quote you an interest rate of 20.9%. You, in turn, would consider this $1,209 in payment to be equivalent not to the 10% expected rate of return that the bank believes it is, but equivalent to a 97% · $1, 209/$1, 000 ≈ 17.3% interest rate.
You would consider the 7.3% to be the extra spread due to friction—but, of course, who says you are right and the bank is wrong?