Saturday, February 5, 2011

Pay Close Attention, Ladies and Gentlemen...

Any sufficiently advanced technology is indistinguishable from magic.

— Arthur C. Clarke


A little while ago, a fellow named Omer Rosen wrote a little piece entitled "Legerdemath" in the Boston Review. Mr. Rosen apparently used to work for Citigroup at the beginning of the Oughts, on its corporate derivatives desk. As a fresh-faced young tyro, Mr. Rosen spent three years selling mostly "mundane interest-rate swaps and Treasury-rate locks" to corporate clients of the bank. Mr. Rosen does not seem to have walked away from this experience with a good feeling about the work he and his colleagues did there.

In fact, his piece created a minor flurry of attention on Twitter and elsewhere, because it paints the activities of Citigroup's derivatives desk in a very unflattering light. Mr. Rosen writes:

Our clients were non-financial corporations, the Deltas and Verizons of the world, which relied on us for advice and education. Our directive was "to help companies decrease and manage their risks." Often we did just that. And often we advised clients to execute trades solely because they presented opportunities for us to profit. In either case, whenever possible we used our superior knowledge to manipulate the pricing of the trade in our favor.

I never heard this arrangement described as a conflict of interest. I learned to think we were simply smarter than the client. For unsophisticated clients, being smarter meant quoting padded rates. For the rest, a bit of "legerdemath" was required. Most brazenly, we taught clients phony math that involved settling Treasury-rate locks by referencing Treasury yields rather than prices.

If a client requested verification of our pricing, we volunteered to fax a time-stamped printout of market data from when the trade was executed. One person talked to the client on the phone while another stood by the computer and repeatedly hit print. The printouts were sorted, and the one showing the most profitable rate for the bank was faxed to the client, regardless of which rate was actually transacted. If a rate for the client's specific trade was not on the printout, we might create rigged conversion spreadsheets for them to use in conjunction with the printout.

Other sources of profit lay in details that clients thought were merely procedural but in actuality affected pricing as well. Once, a client called after his interest-rate swap was completed and asked to change a method of counting days. Unbeknownst to him, this change should have lowered his rate. I made the requested change but kept his rate the same, allowing us to realize unwarranted profit. This was standard practice. My coworkers knew what I had done, as did the traders, as did the people who booked trades. I even tallied the "restructuring" as an achievement in a letter angling for a higher bonus.

Given in what high esteem Wall Street and its employees are currently held in this country and elsewhere, Dear Readers, you can just imagine how this description was received by the chattering classes. Fraud! Deception! Very Naughty Banks Behaving in an Unmistakeably Despicable Fashion! Mr. Rosen's story was taken as almost perfect confirmation of what everybody already knows: that investment banks are completely out for themselves, and that they try (and mostly succeed) to swindle their clients every chance they get.

Sounds good. Too bad that storyline is complete bullshit.

* * *

First, a little background. As I have said many times before, investment banks' businesses can be characterized in one of two ways: agency businesses, where the bank acts on behalf of a client to execute a transaction and earns a fee for doing so, and principal businesses, where the investment bank acts as a counterparty to its customer, and earns a profit or loss on a trade. Traditional investment banking businesses like mergers & acquisitions advisory and securities underwriting fall squarely in the agency camp, whereas full-blown proprietary trading, in-house hedge funds, and in-house private equity funds fall squarely on the side of principal activities. Traditional market-making—where an investment bank stands ready to buy or sell existing securities, commodities, and derivatives for trading customers who want to execute the other side of the trade—can range from relatively riskless agency-type business, where the bank takes on very little inventory or balance sheet exposure to changing market prices, to much more proprietary operations, where the bank hopes and expects to make a profit by holding market positions for more extended periods of time.

But there are hybrid businesses within investment banking, as you might expect from an industry which never discovered a profit-making opportunity it didn't like. One of the most important of these is structured products, in which banks take off-the-shelf and proprietary securities and derivatives and slice, dice, and recombine them into customized instruments that they can then sell to corporations or investors. For corporate customers, these are usually marketed as the solution to some particular asset or liability management problem the company has—like, in the simplest instance, turning a fixed rate borrowing into a floating rate obligation via a fixed-to-floating swap. For (usually institutional) investors, structured products are developed to create a customized or semi-customized investment return tied to various indices, underlying securities or commodities, or almost anything under the sun an investor wants to capture. Investment banks tend to be really good at creating structured products, both because they understand the underlying financial instruments and markets as well or better than anyone and because they have hired a raft of really, really smart propeller heads from academia and elsewhere who can manipulate the complicated maths required to structure them.

But look very, very carefully at the preceding description, O Dearly Beloved, and see if you can detect the pivotal distinction. Did you see it? Did you catch the sleight of hand?

But of course you did, because you are so clever. Structured products are just that: customized products, that are manufactured, marketed, and sold to customers. Now they may in fact be (and usually are) sold as solutions to some problem or opportunity the customer wants to address, but they are products nevertheless. And this bears crucially on the proper understanding of the relationship and obligations between the bank which creates and sells them and the customer which purchases them.

* * *

Think about it. In agency business like M&A or underwriting, the agent-client relationship is clear: the bank is hired by a client to do something, and bears responsibilities and obligations to that client as its trusted advisor. There is no direct conflict of interest with the bank's fiduciary obligations to itself and its owners, since the bank assumes very little real risk in executing such transactions. At the other end of the spectrum, the bank conducts proprietary trades with counterparties. Each party to such a trade knows and expects to know little or nothing about the motivations the other may have or the other's potential profit or loss on the trade; they just bargain for the best possible price they can get.

But in structured products, we are talking about manufactured goods. The bank purchases the raw materials for a trade, creates a structure around these components which delivers a certain advertised set of performance behaviors and characteristics, and sells them to its customer. It sells a product. Accordingly, the customer which purchases a structured product from an investment bank is no more a "client" of that bank—benefiting from a trusted fiduciary advisory relationship—than I am when I buy a can of soda or an automobile. Now it's true that both regulation and norm require that the seller of a manufactured product deliver a good which performs as advertised, and which does not cause unforseen harm or adverse consequences. But Coca Cola or General Motors do not owe me, as the purchaser of their products, any sort of fiduciary obligation or particular duty of care.1

Investment banks are guilty of blurring this distinction themselves, by casually insisting on using the term "client" to describe almost every entity they transact with, whether it is an M&A advisee or the purchaser of a forward start swap. But we are not alone in doing this. Remember the first time you went to look at houses or apartments with a real estate broker, and he or she called you a valued client? Do you also remember how soon you discovered that real estate brokers work for the seller or landlord, who are their real clients, and that you were just a sucker patsy customer? But no matter the particular circumstances, the economic and transactional roles are clear: if someone is selling you a product or a "solution," you are not a client. You are a customer.

And customers do not usually have the right or the ability to see into a seller's manufacturing process to see how the sausage gets made, what type and quality of ingredients are used, and whence the various sources and magnitudes of the resulting profit margin come. I don't know the kind and cost of ingredients that go into my Diet Coke, or the source and cost of the wiring harness in my Range Rover. Do you? Of course not. Manufacturers manufacture things to make a profit. The magnitude and source of that profit is, for me as a consumer, not even a secondary concern. My concern is to get a reasonable quality product which satisfies my needs for the best possible price.

I can try to mitigate the fundamental information asymmetry between the manufacturer of a product and me as buyer in one of two basic ways: I can try to educate myself about the product, and become a more sophisticated consumer, or I can set the seller up in price competition with one or more other sellers to get a better price. Both of these approaches become more difficult for structured products the more customized the product or solution at hand is. And both of these approaches are resisted heartily by the manufacturer, whether it be an investment bank, a soda pop company, or an auto OEM. (Since when did you see any business offer to forgo profits on its activities?) Unless a consumer can manufacture the good she requires herself, she is ineluctably at the mercy of the seller, mitigated only by her negotiating skills.

* * *

So count me among the distinctly underwhelmed and non-outraged over Mr. Rosen's tale of woe. Show me examples of for-profit manufacturers that cheerfully offer their customers complete visibility on all their embedded sources of profit, or who refrain from a little game of hide and seek with customers who press them. The information games Mr. Rosen relates arise exactly because some investment bank customers try to become more sophisticated about their purchases, in order to increase their negotiating leverage over price. Some clients are indeed very sophisticated, and those tend to get the best terms and the best price. Most customers are not; they are price takers. But that is true of any market.

Mr. Rosen seems chagrined that he concealed information and misled clients about his employer's pricing. But where was his obligation to do otherwise? Unless he had a contractual obligation to provide true and transparent pricing to these clients, which would have been extremely unusual, there was none. And don't get your panties in a twist about the amounts involved here, either. We are probably talking about basis points, or fractions of a basis point, here. (A basis point is 1/100th of one percent.) There are enough sophisticated customers and enough price competition among investment banks in the structured product markets to have reduced banks' profit margins on the plain vanilla interest rate swaps and Treasury rate locks Mr. Rosen and his colleagues sold from the percent or two range that existed at the inception of the market to mere basis points today. That is one reason why banks have tried to engineer ever more complicated and sophisticated structured products over time: the profit margins are better.

I strenuously disagree with anyone who contends that apologists like me are trying to "retroactively apply caveat emptor principles" to this corner of the financial markets. They have always been in effect here. It is only the foolish and incompetent customer who did not realize it. I would like some of the self-appointed defenders of Corporate America or the assembled hedge funds and pension fund investors in the structured products market to go toe-to-toe with these alleged "victims" over the terms and price of a moderately structured security or derivative. Most Treasurers or portfolio managers I know would kick their asses six ways from Sunday. These are not dumb, lily-livered creampuffs who buy our stuff.

* * *

Mr. Rosen was a young man, fresh from university, when he ventured into the bowels of Citigroup. Perhaps he became disillusioned at what he saw, like many young people who first encounter the petty concealment and manipulation which lies at the heart of any manufacturing or sales enterprise. But this is not fraud, Ladies and Gentlemen, and this is not something that doesn't happen a million times a day in almost every other manufacturing industry around the globe. This is not even capitalism. This is buying and selling, as it has been done since the first caveman traded his extra spear for a cooking pot.

So there you have it, Ladies and Gentlemen. Presto, change-o! I have turned this smelly old sock into a lovely vase of flowers. You may leave your money at the door. Nice doing business with you.


1 Note that nowhere in his little narrative does Mr. Rosen claim that any of the products his desk sold to customers blew up or acted in any way not as advertised. It would have been rare for that to have happened anyway, given how simple and plain vanilla most of these products were, but note that Mr. Rosen does not claim Citigroup sold customers defective goods.

© 2011 The Epicurean Dealmaker. All rights reserved.