Return to Table of Contents

Chapter 6: Advanced

What is the robbing of a bank, compared with the founding of a bank?

-Bertholt Brecht, The Threepenny Opera

Advanced is probably an overstated description of the topics I intend to cover in this section, but it was the best I could think of. Some of the topics contained here are more difficult than those that preceded them. I cannot think of any way around this. Some things are just harder than others.

Finance and Swaps

One of the ways in which you might come into contact with the finance department is through handling swapped loans. I can hear you already, what's a swapped loan? Recall that the bank pays a floating rate of interest on most of its liabilities (i.e. deposits). If you ever see a breakdown of the deposits in the bank you will notice that about half of the total is in money market accounts which pay interest at a rate that adjusts weekly. Because of this Corus prefers to make floating rate loans, since the interest it receives adjusts just as the interest it pays does. Therefore, whenever the bank makes a fixed rate loan it is exposed to the risk of rates moving up, and receiving less interest on the loan than it pays on the deposits. The bank avoids this risk by entering into swap agreements. Through these agreements the bank agrees to pay some financial institution a fixed stream of payments (which is what the bank is receiving in loan payments) and in return receives a floating rate stream of income. If interest rates stay constant, the swap will result in the bank paying out just as much money as it receives. (More or less, there are of course other complexities, which I will not go into, primarily because I do not understand all of them.) If interest rates go down, the bank will pay out more money than it receives, but the cost of funds (the interest paid on deposits) will go down as well. If rates go up, the bank will receive more than it pays, but will not profit, since the extra money goes to pay the depositors. In this way the bank is assured that the spread it earns on its loans, and hence its income, will remain constant regardless of how rates move. That at least is the basic theory.

Naturally, the finance department keeps careful track of all the fixed rate and swapped loans. Sometimes the bank combines many loans together into what is called a "bucket swap." If a loan has been swapped, either individually or in a bucket, it should have a sticker on its files indicating that the loan is swapped and that the finance department should be notified immediately upon payoff. If a large fixed rate loan pays off the finance department might have to make immediate allowance for it by cancelling a swap agreement, or "unwinding the swap." Naturally, if rates have gone down the other party to the agreement will not allow this to happen without being paid a penalty, since at that moment the other party will be receiving more from the bank than it is paying to it.

Partly for this reason, fixed rate loans often include prepayment penalties. There are two basic models of prepayment penalties: percentage of principal and yield maintenance. The percentage of principal penalty is very simple, and is often applied even to floating rate loans. Example: if a loan has a 3% prepayment penalty, and the loan balance is $100M, then the borrower will have to pay $103M in order to pay off the entire debt and have the mortgage released. A penalty of this type will often decline as the years go by, for instance: 3% in the first year, 2% in the second, 1% in the third, and no penalty thereafter.

Naturally, yield maintenance penalties themselves come in two varieties. Both are based on current treasury rates. All fixed rate loans are priced based on a spread to treasuries. If for instance, the loan is for a term of ten years and the agreed upon spread is 3%, than on the day the loan closes the officer will look up what the 10 year treasury is on that day, add three percent to it, and write that number into the note. The first type of prepayment penalty applies only if the treasury rate on the day of prepayment is less than the treasury rate on the day the loan was booked. You calculate it as follows: (a) calculate the present value of all interest payments due to the bank through maturity discounted at today's treasury rate; (b) calculate the present value of all interest payments the bank would receive if the remaining principal balance was reinvested in treasuries, or at some indicated spread to treasuries, discounted at the treasury rate; subtract (b) from (a).

The second method is simpler to describe and generally costs the borrower more money. It applies whenever the treasury rate is less than the rate of the note. Simply take the present value of all payments due to the bank, including the balloon payment, discounted at today's treasury rate, and subtract today's principal balance. (Of course to figure out the total amount the bank is owed you have to add back the principal balance.) Since the first method is more humane, it has of course been discarded. All new notes use the second method. One final word, the prepayment penalty is never referred to as a penalty. It is always referred to as a prepayment charge.

I should interrupt here for the people who are unfamiliar with the term "present value." Calculating present values is one of the fundamental cornerstones of financial analysis. The concept it captures is that a dollar today is worth more than a dollar tomorrow. And not that a dollar in the hand is worth two in the bush as present value calculations assume risk free transactions. It is easiest to understand in terms of annuities. How much would you pay to receive a fixed payment of $100 every year for the next twenty years? Well, the first $100 payment is worth $100, but the next one is a year away. Since you could have invested the $100 and received some interest on it (or to look at it another way, since inflation will have made it worth less) it must be worth less. If we guess that inflation is going to be 5% a year, or apply a "discount rate" of 5%, that implies the $100 widget you could have bought at the beginning of the year will now cost $105. The purchasing power of your $100 is now down to $100/105, or $95.24. (And not $95.00 as I had hastily written earlier. Thanks to Dan Farkas for pointing out my error.) At the end of the second year, the $95.24 will have declined again to $95.24/105, or $90.70. Continue the calculation until year 20. Then, add up all the payments and you get $1,308.53, which is considerably less than the total amount of payments of $2,000. And of course there is a formula on excel which does all this for you automatically.

Back to our regularly scheduled broadcast. As I said, there is a relationship between the prepayment charge and loan swaps. If a loan is swapped the finance department will choose whether or not to unwind the swap depending on its perspective of the bank's position. When the loan department wants to waive a prepayment charge recognition of the finance department's role in this area will often lead to someone, usually an innocent young credit analyst (as I was once), calling the finance department to ask for permission to waive the charge. The young credit analyst is surprised to learn that the finance department can answer that question without even looking up whether or not the loan is swapped. The answer is simply no. There are a number of ways of explaining why this is true, and I will give only what I consider the simplest. A loan is recorded as an asset on the bank's balance sheet at the amount of money that is disbursed. As the loan amortizes, the value given to this asset is reduced accordingly. This way of looking at it makes a certain kind of accounting sense, but does not capture the true economics of the transaction. What really happens when the bank makes a loan is that it sells the borrower a chunk of capital in exchange for a future cash flow. On the day the loan is booked, the present value of the cash flow discounted at the interest rate is equal to the principal balance. As long as interest rates stay constant, the present value of the cash flow will remain equal to the book value of the loan. However, if interest rates go down, the present value of the cash flow will of course go up. (Think about this until you see why.) So, a borrower trying to pay off a loan for only its principal balance is actually trying to buy an asset from the bank for less than it is worth. The bank is not in the business of selling its assets for less than they are worth, so the finance department will always say no. However, other factors might weigh in favor of waiving the fee. If for instance, the bank can facilitate getting a problem loan off the books, or if an officer needs to give a consideration to an important customer, or etc.

Thinking about swaps, prepayment charges and the finance department is a good way to begin to have a better picture of the bank as a whole and not merely a myopic view of the loan department. Of course down that road lies dissatisfaction with the loan department, unhappiness and boredom in your job, overarching ennui and ultimately suicide. Better to return to thinking about loans as quickly as possible.


I think the most important question someone could be left with after reading this manual this far is: Why would anyone ever get a commercial real estate loan at all? If I have $250M to invest, I'll simply buy a $250M building, and make a nice return of $25M per year. In this way, I'll avoid all the hassles of dealing with bankers, like negotiating terms, legal entanglements, fees, and the danger of losing my property to foreclosure! The answer to this question is leverage. By buying a bigger property than he could afford with only his own money, a borrower hopes to make a greater return on his capital.

A simplistic calculation runs as follows: instead of investing his $250M in a $250M building, the borrower buys a $1 million property with 75% bank financing. Instead of a $25M income, the $1 million property should have an income of $100M. From this the borrower has to pay his debt service, at an interest rate of 8% on $750M, that comes to $60M/year. This means the customer will make an income of $40M/year, instead of $25M. His return on equity (ROE) has risen from 10% to 16%. This dramatic improvement is why investors are willing to take a greater risk by obtaining bank financing.

Now on the other hand consider why the bank invests its deposits in loans that return only 8% when it could simply buy real estate itself and make 10%. It is because the bank is exposed to a much lower level of risk. Remember the bank loan is 75% of the value of the real estate, so if the property drops in price from $1 million to $750M the bank might not lose any money. (Except for the legal costs of foreclosure, and the interest that never gets paid, and the lost time that bank employees spend working on the foreclosure, and the costs of managing and selling the building. And of course the property could drop in value to $500M. The bank is exposed to a lower risk, not no risk.)

Using a small amount of capital to obtain a loan to make a much larger investment is called leverage. When the return on investment is higher than the cost of the debt incurred, thereby increasing your ROE as in the above example, it is called positive leverage. It is easy to understand why projects with positive leverage are attractive to the bank and to the borrower. The bank feels secure that its loan can be repaid, even if the investment does not perform as well as expected, and the borrower can expect to increase his ROE.

Where there is positive leverage there must also be negative leverage, and it has the opposite effects. When the cost of debt is higher than the return on investment everybody loses. The bank is worried that the borrower will have to come out of pocket to meet the debt service and the borrower will see a decrease in his ROE. If you are ever underwriting a deal and determine that it is negatively leveraged you have to think carefully about whether or not you can support the bank entering into that transaction. You also have to be sure you have correctly analyzed the project. The borrower has as little motivation to enter into a negatively leveraged deal as the bank. Assuming the customer is sophisticated and intelligent, what does he know about the deal that you do not? (Tim Stodder and I have a disagreement about this. He argues that if the LTV is good, and the DSCR is good, and the guarantee is good, then the loan is good. If for some reason the borrower believes that his capital is best invested in this way, so be it. I think Tim is full of it on this one, but he's a First Vice President and I'm a Credit Analyst, so I thought I would give you his opinion. Now, as long as I'm in a parenthetical remark, I might as well point out that the ultimate example of a leveraged investment is a bank. A bank starts with a relatively small amount of capital and incurs huge liabilities, i.e. deposits, and uses them to make quite large but low return investments, i.e. loans.)

Determining the return on equity for a construction project can be particularly difficult. The budgets of most projects contain line items of what can only be described as a suspicious nature. Extra developer's fees can be hidden in a number of different line items. It is very possible that the developer's actual return on equity is quite different from what you calculate it to be.

Bringing up construction lending was particularly craven in this instance, since it gives me a chance to segue into one of the final topics I plan to discuss. The essential thing to think about in construction underwriting is the requirements of the take-out lender. (Take-out = repay.) Why? To a certain extent the Bank is concerned with whether or not a project will be profitable, but it is much more concerned with whether or not its loan will be paid off. The bank should never make a construction loan unless it is willing to make a permanent loan on the project. However, since the permanent loan will generally come from some other source, you should always consider what is happening in the lending market and what the actual exit strategy your borrower is likely to employ to pay off the bank's loan. Therefore, one should have some idea what the competition is up to.

The Competition

Take out lenders include other banks, life insurance companies, Wall-Street conduit lenders, and real estate investment trusts (REITs). Since other banks are generally comparable to Corus (although some function like conduits, see below) I will restrict my discussion to the three other competitors.

Life insurance companies are huge players in the commercial mortgage market. In general, they are not interested in the hassles presented by construction lending. Instead they focus on term loans, generally 10 years. Life insurance companies are interested in these types of fixed rate loans, since they meet the requirements of their business: secure investments that produce a steady stream of income.

Wall Street conduit lenders are a very different breed, and a relatively new entrant to the commercial real estate mortgage market. Conduit lenders are also usually only interested in term loans. This is because they are in the business of selling loans. They work by starting with a huge chunk of capital, say $1 billion. They go out and lend it all. In the portfolio of mortgages the conduit accumulates, they try to be sure there is great geographic and property type diversity. The portfolio is then divided, not by the different loans, but into different types of bonds (known as strips). These bonds are called commercial mortgage backed securities (CMBS). The securities are arranged according to risk, and rated by an independent rating agency. The lowest risk bonds are sold at yields only a tiny margin above the comparable United States treasury bills. This is because, in theory, they are so unlikely to go bad, since many properties in the portfolio would have to fail before these bonds could not be repaid. As the riskiness of the bonds increases, the margin to treasuries paid increases proportionately. The final strip is known as the unrated strip, and it pays the highest spread to treasuries, but is the most likely to fail.

Conduit lenders make money by selling the bonds that pay a lower rate of interest than that charged to the borrower. This spread is usually tiny, and they only make money because the size of the bond issues is so enormous. Recently (fall 1998), the CMBS market crashed, and the largest conduit lender, Nomura, which also happens to be the largest originator of commercial mortgages in the country for the past several years, posted a one quarter loss of $1.7 billion (with a b) largely due to the fact that its bonds sold for less than it expected.

Conduit lenders generally make their underwriting standards quite public. Sometimes they will come in terms of a pricing matrix, that shows what pricing they will give a loan based on the property type and the projected DSCR. (Obviously, that's just a tiny bit circular, inasmuch as DSCR depends on loan pricing.) Conduits are much more concerned with DSCR than LTV, since they need to be able to convince people that their bonds will make their scheduled payments. They call mortgage lending "a DSCR game." They are right inasmuch as while the debt service is being covered it's a DSCR game. They are ridiculous inasmuch as the instant the debt service is not met, it becomes an LTV game. I write sneeringly based on the $1.7 billion loss referenced above. (Sneeringly, and illogically too. The loss was caused neither by either poor underwriting nor even by loans going bad, but by a sudden change in the bond markets caused by a flight to quality triggered by the collapse of Asian currency and stock markets. It's the global economy, dude.)

Real estate investment trusts (REITs) are not really competitors to banks. In fact, they are sometimes customers of banks. However, they increase competition by effectively removing properties from the mortgage debt market. REITs are strange entities that are actually closely related to publicly traded corporations. REITs raise money by selling shares on Wall Street and use that capital to buy real estate. The enormous advantage that a REIT has over other corporations is that they do not have to pay taxes. And, as long as they can raise money by selling shares, they have no need for mortgage debt. They do sometimes take loans from banks or other lenders when they believe that leveraging their assets will lead to greater profitability. But, when they do this, they generally do not get real estate loans on their properties. Rather the REIT as a whole will apply for a business loan. This can create an enormous opportunity for the business lenders, but is generally a curse to real estate lenders. REITs have effectively removed billions of dollars of properties from the mortgage lending market.

Two Final Thoughts

There are lots of other topics I could delve into. Letters of credit. Hotel lending with its bizarre talk of RevPARs and rack rates. Mezzanine lending, 1031 exchanges, rent versus buy analysis, and any number of other things. But I have to stop sometime. Any officer can tell you what the above terms mean if any sounds particularly fascinating. Instead, I will restrict myself to two final thoughts.

But first, I actually promised to explain what a letter of credit is. The bank issues a letter of credit at the request of a customer who must guarantee something to a third party who will not take him at his word. For example, company A is requesting some service from company B that will require payment eventually. Company B wants to be sure that it will be paid no matter what. It requests a letter of credit from a bank. Company A comes to the bank and asks for a letter of credit whose beneficiary is company B. Company A promises the bank it will repay the money if the letter is drawn upon. In this way the bank takes away any credit risk company B was facing. Naturally, we charge company A a service for this fee and generally demand some form of security, like a real estate mortgage or a pledged certificate of deposit. On to my two final thoughts.

After working as a credit analyst for a few months, or perhaps after a little longer (and after thoroughly reading and studying this primer!) you might get the mistaken notion that you know what's going on. You'll start to feel as if the senior officers are stifling you; wasting your time and talents on grunt work, when you were obviously meant for the big deals and the weighty issues. You could run this bank better than Bob Glickman with both lobes tied behind your back.

Such arrogant and hubristic thinking is to be fought against. Every single loan officer at this bank knows more about real estate than you do. Every one of them knows better how to underwrite, present, document, service and foreclose loans than you do. Which is not to say that they are infallible. They do make plenty of mistakes and unless the officer you're working with is foolish he'll take your suggestions and criticisms when you're right and present a convincing case. Naturally, none of this applies to me. I'm always right, and deserve your respect and attention at all times. And don't you forget it.

That was the first final thought, now on to the second. The cracks about usury notwithstanding, I am actually unashamed of working at a bank. More or less. Obviously, I could not have worked here for almost three years now if I truly thought charging interest was immoral. On the contrary, I believe that the capital formation and lending that banks do are essential in making an economy work. I think only an extremely literally minded Marxist would disagree with that statement. Banks really perform two very useful functions: they give people someplace safe to put their money and they lend money to people who use it productively. Who can forget Jimmy Stewart's explanation of the savings and loan industry in "It's A Wonderful Life?" Ahhh, the kindly face of capitalism.

Of course banks also have a tremendous power to change our landscape in less positive ways. For every working class stiff who finally gets a home mortgage so he can move out of the slum-lord's building, how many are refused? And, perhaps more importantly, how many slum-lords get their properties financed from that same bank which refuses to give the poor tenants home loans?

One statistic that every commercial real estate loan officer should be aware of is that between 1990 and 1996 the amount of developed land in the Chicago metropolitan area increased by 40%, while the population increased only 9%. Similar statistics for different time periods are equally disproportionate. The people who are against this call it urban sprawl, others call it development. I'd like to think that I'm as for development as the next guy, but I'm also against urban sprawl. I don't support the road building and corresponding increase in traffic (if you build it, they will drive) and corresponding increase in pollution, the loss of farmland and open space, and the fracturing of communities. And of course the appalling ugliness of most new subdivisions is enough to make anyone wince.

Lenders say that they only make the loans, the developers decide where and what to build. That sounds like a reasonable argument. But the developers say that they only build where tenants and buyers are renting and buying. And the business owners say, we're just moving our businesses to where the developers are putting the buildings. And homeowners say we're only trying to get a nice quiet house away from the congestion and crime of the city. And the road builders say, we're just putting the roads where people are driving. And in that way, each does exactly the logical thing, the profitable thing, the self-interested thing. And in that way our beautiful little planet is destroyed.

Perhaps the lenders are right and there is nothing they alone can do. Of course perhaps the owners, and the developers, and the tenants, and the home-buyers are right and there is nothing that can be done. I hope not. I hope that lenders can play a part in a real solution to the problem. Or perhaps we will all be saved when the glorious day comes that sees us all in a paradise where money and property are done away with and our problems simply dissolve away. I'm not holding my breath though. In 1649, Gerrard Winstanley told the real estate mavens

the power of enclosing land and owning property was brought into the creation by your ancestors by the sword; which first did murder their fellow creatures, men, and after plunder or steal away their land, and left this land successively to you, their children. And therefore, though you did not kill or thieve, yet you hold that cursed thing in your hand by the power of the sword; and so you justify the wicked deed of your fathers, and that sin of your fathers shall be visited upon the head of you and your children to the third and fourth generation, and longer too, till your bloody and thieving power be rooted out of the land.

Still waiting.

Return to Chapter 5: Procedures