I am endlessly fascinated by the post-banking career of Greg Smith, the former Goldman Sachs derivatives salesman who left to write a book about being a Goldman Sachs derivatives salesman, though that's partly for idiosyncratic reasons. I'm a lot like Smith, you see. Each of us worked at Goldman, at about the same time, each ending up as a vice president, each hawking equity derivatives, albeit different derivatives to different clients.* And we each gave up those jobs for the more precarious business of writing about the financial industry. It is a quiet bond we share, though unlike Smith my Goldman career did not involve throwing a Ping-Pong tournament to please a client, so I can never fully imagine what it is like to be him.
Anyway, today we learned that he met with the Securities and Exchange Commission a few weeks ago to tell them how to write the Volcker rule, which is really just delightful. Writing the Volcker rule is hard, says everyone. And who better to tell you how to write the Volcker rule than a disgruntled former midlevel derivatives salesman? Lots of people, probably! And, to be fair, the SEC is meeting with lots of other people to talk about the Volcker rule. How could they turn down the meeting with Smith? I am reliably informed that SEC staffers enjoy a laugh now and then, too.
Oh, I'm being unfair. You can read the agenda for Smith's meeting here, and it's actually a useful perspective and one that -- since I'm basically Smith's evil twin -- I sort of share.** One of the big debates in the Volcker rule drafting is over the term "market making." Regulators want to ban "proprietary trading" by banks, which means trading by banks for their own account, but they want to allow "market making," which means banks offering to trade (for their own account) with clients in order to facilitate clients' desired trades. The idea is that a bank shouldn't go around trying to buy lots of shares of Facebook stock or whatever because it thinks that Facebook is a good investment, but if a client wants to sell Facebook stock because she thinks it's no longer a good investment, her bank should be able to buy it from her.
This is not an entirely coherent idea, but it's the idea, so whatever.***
Here is Smith, from the agenda:
Market making should be when a customer approaches a bank to facilitate a trade. Today, the bank decides what trade it wants to do -- then it lines up clients to take the other side of the trade. This is no longer market making in my view -- it is akin to proprietary trading.
Doesn't that just sound right? If the client wants it, it's market making. If the bank wants it, it's prop.
So: No, it's not right! Market making is doing trades with clients where they buy from you above the fair price and sell to you below the fair price because they are paying you: You are providing them some useful product ("liquidity," I suppose, or in derivatives maybe "hedging" or "a tax dodge" or whatever). Prop trading is doing trades with other people where (you hope) they buy from you above the fair price and sell to you below the fair price because you are smarter than the market and you know what the fair price is and the market doesn't.
That's a pretty squicky distinction really, because market makers also want to be smarter than, and tend to know more about fair prices than, their customers. A useful way of putting it -- and one used by the SEC, among many other indications, in its own proposed definition of market making -- is that a market maker gets paid the bid-ask spread, while a prop trader pays the spread. If you're getting paid to trade, you're market making; if you're paying to trade, you're prop trading. But this is not an infallible rule, particularly in products where there's no obvious market price and so no easily observable spread. Market making is buying low and selling high, but so is everything really.
Anyway, though, if your business is getting paid to trade you should want to trade more and so you should spend a lot of time going around to clients and trying to get them to trade with you. You should also spend a lot of time trying to build new products that they'll want to trade with you, because those products provide the client some advantage that no one else offers. Also, though: You should spend a lot of time trying to build new products that make you a lot of money, i.e. that are mispriced in your favor, because you want money, and less transparent products can build in a bigger bid-ask spread and make more money off clients. "A bigger bid-ask spread," for a derivative valued on an expected-value basis using a theoretical model, is not entirely different from just "you do trades you think are undervalued." Which sounds sorta prop!
True story: Like Smith, I worked on a desk where the job was to dream up derivative trades in which Goldman would put its own money at risk in long-term, client-facing trades, make sure they were profitable (on an expected value basis), and then go out and find clients to do them with. When rumblings about the Volcker rule started, I just assumed that I was in a "prop" business. Nope! But the fact that it's a "market making" business -- even though it was a distinctly unusual event for a client to call us out of the blue and say "hey could you make me a two-sided market on some crazy derivative?"**** -- was not at all intuitive to me. Or to Smith. Like he said: it's not market making, in his view.
But that's just his view and it's wrong. I mean it's wrong as the market uses the term, and it's wrong in the sense that, if the SEC adopted it, it would shut down pretty much all Wall Street market-making. Wall Street firms do not make the bulk of their money by sitting quietly and waiting for the phone to ring with clients asking them to buy and sell some stocks and bonds. There is a reason they hire salesmen and treat clients to rigged Ping-Pong tournaments. Those derivatives don't sell themselves.
Of course if the SEC did shut down all derivatives sales desks Smith probably wouldn't be too broken up about it. (He's made his money selling derivatives!) And I guess he wouldn't be alone. It's a weird view -- how many industries are there where companies can sell a product only if the customer asks for it unprompted?***** -- but also a weirdly popular one, and one that Smith is well suited to present. And the fact that the SEC is ... unlikely to take him too seriously is, from my perspective, a nice bonus.
* As far as I know we never met.
** But, y'know. In an evil way.
*** Perhaps I'm being unfair on its incoherence; the rule's intuition that market-making is less risky than prop trading is a fairly reasonable one. I tend to think that, if you like the Volcker rule, you should love the more coherent Glass-Steagall-ish idea of separating insured banking from all sorts of trading activity, prop or market-making or otherwise, but that's not especially on the table. (Also I'm oversimplifying in the text, obviously: if a bank wants to buy Facebook as an investment that's actually fine! They just can't buy it as a short-term investment: They have to plan to hold it for at least 60 days. Shorter than that is prop trading for some reason.)
**** No, I mean, I worked in a business where two-sided markets were not a thing, because the clients were sort of legally constrained to one side of each product, so that basically happened zero times. But clients did occasionally call us out of the blue and ask for a price. But for the most part those products were sold, not bought, as they say.
***** Tobacco, right? Probably some others. Again you could imagine people lumping derivatives in that category.