Inside Goldman’s ABACUS Trade

Today, in the Huffington Post, I posted a document that shows an earlier incarnation of the ABACUS trade (although, not that different from the one that has got the SEC up in arms). I also explained it as well as I could. Head on over and let me know what you think.

Why Stress Test Really Means Guesswork

Well, we’ve heard a ton about stress tests recently. Want some details on what a stress test entails? The Journal has some details about the tests here. Now, as much as I think GDP and unemployment are fine things to project forward for economists, let’s walk through the way one would use this to actually price an asset. Let’s start with something simple, like a 10-year treasury note (note that treasury bond specifically refers to bonds with a 30-year maturity). Here are all the components one would need to stress test the value of a treasury note.

1. Characteristics of the note itself: coupon, payment dates, maturity dates, etc.
2. What the yield curve would look like at the date you’re pricing the note.

Why would one need to know the shape of the yield curve (term structure of rates)? This is important, in order to “PV” the bond’s cashflows most accurately, one would discount each cashflow by it’s risk–the simplest proxy is to discount each cash flow by the rate of interest one would need to pay to issue a bond maturing on that date. For the government, this rate of interest is the point on the treasury yield curve (actually, the par zero curve) with the same maturity date. An example would be, if I were going to price a cash payment I will receive in two years, and the government can currently issue two-year debt at 5%, I should discount my cash payment (also from the government, since it’s a treasury note) at 5%. Treasuries are the simplest of all instruments to value.

Here’s an example, form the link above, of what a treasury yield curve might look like:

Now, it is completely and totally guesswork to figure out, given unemployment and GDP figures, what the yield curve will look like at any date in the future. Indeed, one can plug these projections into a model and it can come up with a statistical guess… But the only thing we know for sure about that guess is that it won’t be accurate, although it might be close. However, things like inflation will drive the longer end of the yield curve and monetary policy will drive the shorter end, so these certainly aren’t directly taken from the stress test parameters, but would need to be a guess based on those parameters. This is a large source of uncertainty in pricing even these instruments in the stress test.

Next, let’s examine a corporate bond. What would we need for a corporate bond?

1. Characteristics of the bond: coupon, payment dates, maturity dates, special features (coupon steps, sinking funds, call schedules, etc.), where in the capital structure this bond sits, etc.
2. What the yield curve would look like at the date you’re pricing the bond.
3. The spreads that the corporation’s debt will carry at the date you’re pricing the bond.

Oh no. We already saw the issues with #2, but now we have #3. What will this corporations credit spread (interest/yield required in excess of the risk free rate) at the time of pricing? Will the corporations debt, which could trade at a spread of anywhere from 5 to 1500 basis points, be lower? higher? Will the corporations spread curve be flatter? steeper?

Here is a good illustration of what I’m referring to (from the same source as the figure above):

There, the spread is the difference between the purple line and the black line. As you can see, it’s different for different maturity corporate bonds (which makes sense, because if a company defaults in year two, it’ll also default on it’s three year debt.. but the companies’ two year debt might never default, but the company might default during it’s third year, creating more risk for three year bonds issued by that company than two year bonds). It shouldn’t be a surprise, after our exercise above, to learn that the best way to compute the price of a corporate bond is to discount each cashflow by it’s risk (in my example above, regardless of whether the company defaults in year two or year three, the interest payments from both the three year and two year debt that are paid in one year have the same risk).

Well, how does one predict the structure of credit spreads in the future? Here’s a hint: models. Interest rates, however, are an input to this model, since the cost of a firm’s borrowing is an important input to figuring out a corporation’s cashflow and, by extension, creditworthiness. So now we have not only a flawed interest rate projection, but we have a projection of corporate risk that, in addition to being flawed itself, takes our other flawed projections as an input! Understanding model error yet? Oh, and yes unemployment and the health of the economy will be inputs to the model that spits out our guess for credit spreads in the future as well.

Next stop on the crazy train, mortgage products! What does one need to project prices for mortgage products?

1. Characteristics of the bond: coupon, payment dates, maturity dates, structure of the underlying securitization (how does cash get assigned in the event of a default, prepayment, etc.), etc.
2. What the yield curve would look like at the date you’re pricing the bond.
3. The spreads that the debt will carry at the date you’re pricing the bond.
4. What prepayments will have occurred by the date you’re pricing the bond and what prepayments will occur in the future, including when each will occur.
5. What defaults will have occurred by the date you’re pricing the bond and what defaults will occur in the future, including when each will occur.

Oh crap. We’ve covered #1-3. But, look at #4 and #5 … To price a mortgage bond, one needs to be able to project out, over the life of the bond, prepayments and defaults. Each is driven bydifferent variables and each happens in different timeframes. Guess how each projection is arrived at? Models! What are the inputs to these models? Well, interest rates (ones ability to refinance depends on where rates are at the time) over a long period of time (keep in mind that you need rates over time, having rates at 5% in three years is completely different if rates where 1% or 15% for the three years before). General economic health, including regional (or more local) unemployment rates (if the south has a spike in unemployment, but the rest of the country sees a slight decrease, you’ll likely see defaults increase). And a myrid of other variables can be tossed in for good measure. So now we have two more models, driven by our flawed interest rate projections, flawed credit projections (ones ability to refinance is driven by their mortgage rate, which is some benchmark interest rate [treasuries here] plus some spread, from #3), and the unemployment and GDP projections.

I will, at this point, decline to talk about pricing C.D.O.’s … Just understand, however, that C.D.O.’s are portfolios of corporate and mortgage bonds, so they are a full extra order of magnitude more complex. Is it clear, now, why these stress tests, as they seem to be defined, aren’t all that specific, and potentially not all that useful?

Detailed Causes of the Crisis and Post-Crisis

Since this is a political season, and with the economic crisis, I think everyone in finance understands there is a sort of “silly season” that ensues. We certainly noted the sort of irrational behavior that would immediately make an economist question their beliefs. To me, though, the most offensive form of this stupidity comes from those who believe the Community Reinvestment Act and Fannie and Freddie sparked the whole crisis. Mr. Ritholtz rails against this notion over and over again. Oddly, I haven’t seen anyone else tackle this issue… Of course, I’m also way behind on reading my feeds. I even wrote Mr. Ritholtz an email (something I always tell myself is useless afterwards, since I don’t ever get a response, but is usually cathartic) noting that he was being very informative by setting the record straight. Well, maybe I expressed this sentiment with a tirade…

Every time I hear a Republican talking head on a news program saying Fannie and Freddie caused the problem I want to jump through my T.V., explain that the answer “betrays not even a modest understanding of the contributing factors to the current crisis, it’s scope, and magnitude” and begin to rattle off about flawed ratings agencies, excessive leverage (for investment banks and funds), over-reliance on models, a flawed compensation model for Wall St., managements needs to one-up their own earnings and those of competitors, explosive year over year growth of unproven financial technologies, over-reliance on “fast money” to distribute risk, fund’s need to earn outsized returns to attract assets, funds’ need to buy crappy bonds to build a “relationship” that would allow them to get “good” bonds from banks, poor disclosure from companies (specifically investment banks, as I’ve discussed on my blog), and extremely low rates for a very long time. Of course I’m just a normal guy who actually knows what’s going on, I don’t get invited onto these shows.

(Emphasis added, mine.)

Let’s tackle these, shall we?

1. Excessive Leverage — If the plot of the credit crisis had included a deus ex machina it would have been an instant de-levering of troubled investment firms. This didn’t happen and several collapsed. I don’t want to be repetitive, but the Deal Professor says it plainly when he says, “Sometimes, You Can Only Raise Capital When You Don’t Need It” … If a firm is highly levered, as Bear was, Lehman was, Fannie was, Freddie was, and A.I.G. was, then when the market gets bad, losses pile up, and credit tightens it’s a death spiral. There’s a large distance between well-capitalized and insolvent, but once you move from adequately-capitalized to under-capitalized it’s probably impossible not to hit insolvent or bankrupt. Oh, and let’s not forget how this became a problem in the first place … the rules were relaxed in 2004.
2. Flawed Rating Agencies — This is pretty obvious. Moodys errors. Rating agencies noting any deal, even one “structured by cows,” would be rated. And lastly, the smoking gun that seems to be the largest caliber, the fact that … well, I’ll let Mr. Raiter speak for himself:… “Mr. Raiter said that the residential mortgage rating group at S.& P. had captured the largest market share among its main competitors — 92 percent or better — ‘and improving the model would not add to S.& P.’s revenues.‘” Wow! Honesty, stupidity, incompetence … all on display. Now, to be honest, I have no idea what difference these problems made. What I do know is that the rating agencies were used as a means of outsourcing risk management and credit analysis. While it shouldn’t be a huge shock that the rating agencies missed the mark, the magnitude by which they missed is a huge problem if everyone took their ratings as fundamentally true. What these “statistical rating agencies” should have been doing is running securities and mortgage loans through abhorrently conservative scenarios and fixing ratings based on those…. they didn’t. They were argued down to “realistic” scenarios based on past experience. The issues above merely compound the problem.
3. Over-Reliance on Models — Related to the rating agencies’ issues, this one is a great catchall for terrible risk management. Let’s be honest, no one saw the fundamentals in housing getting so bad. That’s not the issue, I didn’t see it so I can’t exactly blame others for not seeing it. What I can do, however, is blame risk management professionals for not preparing for it. When you have, as Citi did, tens of billions of dollars in highly correlated assets, you should know there’s a risk of tens of billions of dollars in writedowns. When you have tens of billions of dollars in commercial mortgages, as Lehman did, you should realize the risks there. Similar lessons for WaMu, Wachovia, and CountryWide. Instead, though, like the rating agencies, there was a push to have “realistic” or “back tested” results. Let’s go to Mr. Viniar, C.F.O. of Goldman, for his take: “Even scenario analysis, which can address some of VAR’s deficiencies, came up short … [This] ’caused us to look at more-extreme scenarios than we used to look at,’ says Viniar. ‘It made us expand out the tails of what we deemed a realistic possibility.'” Logical, concise, and conservative. It seems Goldman didn’t attempt to show lower risk numbers so that they could deploy more capital or be looked upon as safer by the stock market. No, they looked at more extreme scenarios. They reacted quickly. However, in quoting this passage I sandbagged you, dear readers. This quotation is actually much more relevant to this situation than one would think–it comes from 2001! Mr. Viniar, people probably won’t remember (seems like a lifetime ago), but I noted before, was the guy who convened a firm-wide meeting on exposure to the housing markets. The takeaway is that the firm that looked at the most extreme scenarios, not the ones that models said were most likely, weathered the storm the best.
4. Flawed Compensation Model — This one is pretty obvious. Lots of money flowed into people’s P.A.’s (that’s “personal account”) each year based on fees and mark-to-market gains for complex structured products. In many instances these risks were distributed and off the balance sheets of investment banks. However, these businesses were grown, and none of the risks were well understood–the people in the lead, though, lead the charge to increase their compensation. I was personally aware of a senior trader/banker/whatever that pushed a firm, one that has seen tens of billions in writedowns and may or may not still be alive, who pushed for balance sheet commitment of 2-3x the current size in the C.D.O. business. This would have exposed this institution to writedowns larger than most firms equity base. This proposal was shot down, but still… Clearly making eight digits was going to someone’s head. Now, we all know that I believe one should be accountable for their decisions, so it shouldn’t be a surprise that when one has made tens of millions of dollars in bonus and salary, but their decisions lead an institution to take massive losses, reduces shareholder value significantly (keeping in mind shareholders might be woefully unaware of the risks being taken), and leads to thousands of people losing their jobs, merely being fired isn’t enough. Especially since these issues are only beginning to be understood when these people are fired, usually. Becoming an instant millionaire is a huge, huge problem. It’s the “swing for the fences if you’re down” mentality, and it’s also the “worry about the tail events if they happen” mentality. Put simply, there should be the ability to claw-back compensation based on performance for years. Perhaps a ten year lockup of wealth is extreme, but given these issues and famous blowups in the past, and taking into account the tradition of good times to last several years, maybe ten years is harsh but not extreme. Maybe employees should be allowed to hedge exposure to stock prices after a few years, but still have risk if negligence is discovered or things go wrong that were set in motion by that person. Obviously something drastic needs to be done, perhaps merely paying less is sufficient, but I doubt it.
5. Management Pressures — Highly correlated to the flawed compensation model, it’s the case that management was pushed hard to get earnings up. Having seen the “budget” process (an odd name, I thought, since a budget, to me, merely means expenditures) up close, I saw people come up with reasonable numbers, submit them to senior management, and be told, “More!” Well, guess what <expletive>s? If someone tells you they can reasonably deliver something and you always add 10-20% to those numbers, there is more risk taking and less rationality to how that profit is achieved. Maybe the long term effects of pushing the envelope are much worse than not taking those risks to begin with. This is one reason Goldman seems to outperform so often, they understand what they are getting themselves into. They truly work together and achieve revenues through teamwork instead of edict. Now, underperformance is punished, but setting reasonable goals is step one when trying to exceed them. The next generation of management should fight their bosses tooth and nail not to set unreasonable baseline expectations and should figure out objective measures that reflect an employee or business’s effectiveness. The tyranny of quarterly earnings shouldn’t make grown ups act stupidly because they can’t “just say no.” Here’s a hint: if you run a company with a nine- or ten-digit balance sheet and you don’t realize your business is complex enough that you shouldn’t manage to the next ninety days, then you should step aside. Seems simple to me. Maybe that’s why Google doesn’t bother with quarterly guidance.
6. Explosive Growth of Unproven Financial Technologies — Being a bit of a purist I am hesitant to call financial products or methods “technologies,” but I’ll use that word for now. The truth of the matter is, these products had never seen a massive downturn. Sub-prime loans as we know them today hadn’t seen a recession until now. C.D.O.’s backed by structured products hadn’t existed during a protracted period of fundamnetal credit distress before. This was known and talked about often. For as much as this was talked about, it was an observation that was never extrapolated. Hedging and risk management still looked at historical levels of distress and credit problems. The market had grown by orders of magnitude, but that wasn’t part of the equation. Quite simply, the fact that these markets grew so much so fast meant no one had a good handle on the feedback effects of this growth. This is somewhat obvious and very moot, so I won’t dwell on the problems of such massive growth.
7. Over Reliance on “Fast Money” To Distribute Risk — Anyone who knows structured products understand this point. Basically, the fair-weather buyers are “fast money.” This client based is distinct from “buy and hold” or “real money” accounts. Here is where the shell game of wall street really kicked into high gear. Hedge funds would buy bonds with the intention of selling at a profit later. Investment banks would, to show strength of the market, put out “bids” or interest to purchase securities they had just created at a higher price than they had just sold said securities at. Hedge funds would then immediately sell back to Wall St. firms, at a profit, to take advantage of their desire to show the market their securitizations “trade well” or “at a premium.” When firms are making money on the securitization, they can afford this. Speaking more generally, hedge funds just “trade bonds around” more. In recent years insurance companies and banks, the institutions that buy securities and rarely sell them (for a myriad of reasons), went from 70+% of the buying base for structured products to 20-30% of the buying base. This means that in a bad market 70-80% of the bonds that exist can be sold (dumped?) at a moments notice. Add in the fact that during this period there was explosive growth (as noted above) and you see why when the markets hit trouble the huge wave of selling occurred, liquidity dried up, and prices plummeted.
8. The Flawed Model for Relationships Funds have with Wall St. (coupled with Funds’ Needs for High Returns to attract Assets) — The way a bank figured out if a hedge fund was a good customer was, basically, how much a fund helped that bank get out of risk (stupidly, as stated above, since banks were likely to be more hurt by a fund owning assets and were more likely to wind up needing to repurchase those assets, but I digress…). However, when assets were in short supply relative to demand, only the top clients were able to purchase securities banks were creating. So, one might wonder, how did a nascent fund, at the bottom of the food chain, get access to the desirable securities? Easy solution: they purchased the undesirable securities to “help out” a Wall St. firm. These were more risky, although they were generally carried a higher rate of return in the event of no credit problems. These new funds, then, showed higher returns, attracted more money, and bought more securities from banks. Net effect? Most funds had a poor mix of products–higher risk bonds or assets that would get hit much harder than generic securities and more generic securities. Keep in mind that, to get high returns, funds were buying C.D.O. products and other structured products that had higher returns in general, but funds also levered these products and thus funds were much more exposed to moves in the market. Funds, as everyone knows, get paid a percentage of assets under management and returns, so to grow their revenue stream many funds just had to buy lots of securities (and, to establish a strong enough relationship to be allocated enough securities, plenty of lower quality securities). This was the prisoner’s dilemma of the syndicate system–funds cooperated every time. (Just to put some numbers on it, when a fund would try to buy residential or commercial mortgage backed securities it was possible for demand to outstrip supply 2- or 3-to-1. Accounts with strong relationships usually got 100% to 80% of the requested amount of bonds being issued. Weak relationships or smaller firms could receive as little as 10-20% of their desired allocation.) This is a complex process and nuanced point, feel free to email me for more explanation.
9. Poor Disclosure from Companies — This is a point I’ve raised before. I won’t go over it again. The short story is that firms got away with a lot because they didn’t tell anyone what they were doing.
10. Extremely Low Rates for a Very Long Time — I’ve raised this point before as well (between the numbered lists). Rates were very low and, suddenly, a product that trades at 50-100bps over L.I.B.O.R. traded 50-100% higher than L.I.B.O.R. If your benchmark was treasury rates to outperform your benchmark meaningfully you needed to get much higher spreads, and thus take higher risk. This is why C.D.O.’s experienced such explosive growth (see the problems the growth cased above). Low rates also made it more attractive to get a floating rate mortgage, which a huge majority of sub-prime mortgages were. This was part of the ex-post concern with Alan Greenspan’s encouraging people to take out A.R.M.’s.

In short, Fannie and Freddie were part of the problem, but not in and of themselves. In fact, if Fannie and Freddie had caused these problems by selling banks their bonds, then we wouldn’t have a problem at all. Why? Because Fannie and Freddie would be “on the hook” for the bonds they guarantee. If these bonds went bad no firms would have taken losses on them (since the government stepped in to keep them solvent and backstopped their obligations). Okay, now that I’m done ranting I’m going to rant on something new. The post-crisis narrative of what went wrong… (don’t you love the rise of the word “narrative”?).

1. The failure of rating agencies, risk managers, and risk management models. This has been getting the most press because it’s easy to explain (not why these things failed, but the fact they failed).
2. Sheer size. This is pretty silly, if you ask me. Bigger doesn’t have to mean riskier. The practices that get a firm to a massive size could be an issue. Super-concentrating the health of the markets with very few players could be a huge problem. The “Too Big to Fail” issue might fit some situations, but didn’t cause this crisis. No one wants to have to rely on the government to save them.
3. Executive pay. This is a limited view on the actual problem. In fact, in most firms, C.E.O.’s aren’t the highest earning individuals.
4. Hedge funds and short selling. Really? Let’s trace the logic here (or lack thereof): a firm runs it’s business poorly and I bet it will decline in value. Clearly I am at fault there. The “free markets at all costs except losses” crowd, like those currently at Treasury, are putting a band aid on an amputated leg here. Especially with the very firms begging to be protected turning around and getting fees from products circumventing the bank on short selling. (What a stupid move, some firms deserve to be in worse trouble.)
5. Everything else. Why get into the details of the actual causes when you can distill down issues to “good” versus “bad” and simple fights? No one has…. so I’m doing it! But I doubt all the other things will make it into the popular understanding of what went wrong.

There you go. My hands are tired, so I’ll stop here. Feel free to comment and ask questions.

From the Ministry of Obvious

Okay, I’m a huge fan of Research Recap, so don’t take this as “ragging” on them, but I was reading my feeds and came across this gem of a headline:

Synthetic CDO Issuance Down Sharply in First Quarter

Wow! Really? Apparently CreditSights, whom I have heard good things about, put out a report saying this. The money quote? Here it is…

The Cash flow CDOs that are being launched increasingly appear to be designed to help banks clean up their balance sheets rather than attempts to arbitrage the agency ratings.

And then there was this other gem in the post, quoting the report…

“Any widening, it was claimed, would rapidly be exploited by a wave of CDO issuance. The most important driver of this stabilisation was synthetic CDOs – specifically the idea that bespoke single-tranche deals could be placed with investors without the need to fill the entire capital structure and this protection selling would push spreads lower.”“Such arguments have been demolished by the events in the past 12 months with both synthetic and cash flow CDO issuance falling like a rock owing to a slew of economic, ratings, and funding concerns.”

So why is CreditSights (CreditHindSights, in this instance) releasing such a report, detailing what everyone with a minimal attention span and the ability to read a newspaper would be able to figure out for themselves? Oh, right …

The full report is available for purchase.

(link omitted)

I don’t think a firm needs to sell a report telling people interested in reading finance research that people aren’t buying CDOs like they used to, anymore.

Maybe their next report will analyze Bear Stearns most recent 10-K and detail some warning signs they see as troubling…

Build an Investment Bank: Technology

(As the first in this series, I’m trying to use construction terms to “build” our investment bank… we’ll see if it adds or detracts.)

The Foundation

As we begin our journey to build our very own investment bank, I’m going to make a few statements that people “in the know” will find both surprising and, in hindsight, very obvious. The topic, as the title states, is technology. Now, here are the statements:

1. A major contributing factor to the way banks did business, especially in the businesses that contributed the most to banks’ current  problems, was their lack of technology.
2. Credit default swaps, in all their glory, had most of their issues rooted in technological inadequacies at various institutions.
3. A large portion of the cost structures at investment banks are due to a lack of technological heft.

I know, these seem outrageous. However, as anyone who has worked at a few different firms will tell you, there is a massive difference between a firm with good technology and bad technology. Let me tell you a simple anecdote: When very senior executives at a firm called down to the managers in charge of securitized products, they asked for the current marks and a summary of the various exposures “on the books.” It took about ten people three days to cull through all the various positions, put marks on them, model them, and put a concrete value on them. There wasn’t time to break down exposures by anything but the most trivial categories. Now, why this end product was acceptable is a different issue, but it should be clear that an effort of this magnitude shouldn’t be necessary to answer questions so totally basic in the context of running a multi-billion-dollar (although now with fewer billions) financial institution. A corollary: If it takes you several days to enumerate the positions your area has, you don’t know what it is yourself.

Now, when I speak of technology, I’m really speaking of the specialized systems and solutions used to tackle business issues, and not really the “desktop support” kind of technology. The systems that manage risk and positions, handle accounting, maintain an integrated analytics platform, deliver research and other products internally and externally, manage the human resource functions of the firm, and otherwise grease the wheels of capitalism.

The Blueprint

Our technology plan will have a few different components…

Structural Frame 1: Whether our theoretical investment bank is a startup or an established entity, the technology at the core will be home grown.

Structural Notes: Hiring consultants to stitch together purchased solutions and legacy systems is unacceptable. Technology, in order to be most effective, needs to be responsive. When a trading desk needs to run its business, and the system provided is insufficient, then it’s an unacceptable solution, and things will be done manually. Remember synthetic CDOs? Remember the ABX and credit default swaps on sub-prime bonds? Would it surprise you to know that at many major investment banks there was a manual component involved with every single contract and trade? The systems weren’t able to handle these instruments, and these businesses scaled up at a rate that was untenable. Also, there were no analytics available for these products. Businesses bought third party solutions for modeling and analytics, but those didn’t integrate or scale, so all the marks and risk numbers used to compute capital needs and P&L were merely estimates as these businesses were growing the most.

Let that sink in. Is it any wonder the senior managers didn’t know, before it was too late, what the actual exposures were? Had these firms built an integrated set of systems instead of buying a patchwork of specialized programs to solve the most current problem, these issues would not have been nearly as bad. I won’t even tell, in detail, the story about how, years ago, the system for trading credit default swaps at one bank was so difficult to use that they only created one identifier for GM and GMAC, not distinguishing between the two at all. But, when they were both on the brink of being downgraded to junk, but GMAC was de-coupled from GM, I wonder what kind of fun it was to rummage through 5- to 8-year-old confirms trying to match thousands and thousands of trades with the exact entity? Costly? Absolutely. Avoidable? Double absolutely.

On another note, an investment bank need not be innovative, but if it isn’t, then it should be able to mimic innovations quickly. Reporting to management, having an accurate record of transactions and various changes to the firm’s balance sheet, the ability to run various analyses on various products, and other, more basic, reporting functions (not even mentioning compliance and regulatory functionality) are all things that should be implementable once something new hits, and the only way to make these kinds of incremental changes is to build, not buy. A business as complex as an investment bank shouldn’t be reliant on outside parties to build software vital to their business–both from a cost standpoint and from a delay-until-completion standpoint. Further, the procurement process takes months!

Structural Frame 2: The technology part of the organization will not be a monolithic standalone bureaucracy.

Structural Notes: Simply put, technology (the people or business unit) needs to be vested in the process of making a business more profitable. Rather than taking on the normal support role mentality of, “If I say ‘yes’ then I might be wrong and held accountable, so I will say ‘no.'” The best way to do this is to not have technology be its own portion of the organization. Allowing technology to have a seperate seat at the table–or, worse, report into some catchall support person–only contributes to creating a centralized process for technology decisions. Centralizing technology decisions for many businesses with different needs creates unnecessary layering and wedges a huge management structure between the people doing the actual work and the people who are using the product and paying for it.

The final plan, I believe, would be to have as many technology people as possible integrated into the physical workspace of the people that utilize their work. Have investment banking developers sitting amongst investment bankers. Have the developers that build trading applications sitting with traders. The reporting structure should be a matrix of sorts–senior technology managers should report into a business whose technology needs are distinct from other businesses (atomic, perhaps is a better word) as well as a more senior technology person. In essence, people working in technology would be ingrained with the thought that they are there to help–the business unit would be setup as the client and the technology super-structure would be more for managing the processes. Obviously when the business is viewed as the client, technology managers are incentivized to get the businesses what they want, and when the people (both doing the work and in charge of liaising with the clients/business) are integrated (and can see the working environment of their clients and usage of their products) a lot of inefficiency and “lost in translation” moments are avoided. Senior managers really need to think of their business as including technology instead of interacting with it. This is highly important and is much more likely with a structure like I’ve proposed. Also, the closeness will just yield some more technologically savvy people and even encourage people to move between the two “worlds.”

Structural Frame 3: The people who are hired for technology roles will be of a high caliber and will be under a compensation regime and in an environment that sets big technology companies to shame.

Structural Notes: This shouldn’t be a hard line of reasoning to follow, but in general the difference between firms that “get it” and firms that don’t is how they recruit. Having an engineering background, I was recruited for I.T. from a very good school for that sort of thing by a few banks. Those banks have a high correlation to both still being around and surviving the mortgage mess with the smallest scathing in their peer group. I know several people who have told me that some other firms, one that haven’t been so lucky, have absolutely ridiculous and incredibly stupid policies for recruiting technology people. Most notably, one Manhattan firm recruits from local state schools almost exclusively–this is done so that the students they recruit can work part time during their senior year of college. No school in the top fifty or so participates. If one had to draw a grid, and rank various factors as to how important they are, the program I have just mentioned is the most ridiculous, stupid, and demonstrating a complete lack of critical thinking skills (or, for that matter, basic grasp of the business and reality) of the programs I have heard of or encountered. The people responsible for it have all moved on and the firm has suffered greatly from it’s underinvestment in technology.

So, to recruit good people you need a draw. To be honest, most graduates don’t fully grasp the concept of upside or career path–especially not ones in I.T. This makes it simple to get them, just offer a bigger number for the compensation in the first year. While this would work, it should be clear that this won’t help make them much more productive than the average technology drone in an investment bank. Giving technology employees a compensation structure that matches the businesses they are supporting is, in my view, a great solution. Obviously there would be more stability, but there should be a linking of incentives to the business and an interconnectedness in how they think about how technology and the problems facing the business. They should also have an incentive to be proactive and try to advocate solutions to problems they see instead of waiting for others to focus on them–this contributes greatly to becoming a nimble organization.

As for work environment, whenever possible, for groups not truly linked to a single business, like infrastructure groups and the web development team, my focus would be on building a start up-like atmosphere. The marginal cost of things like free coffee, free food, and some extra square footage for odd amenities is insignificant in relation to the quality of the work produced by the people snatched from places like Microsoft and Google versus a lower caliber of student culled from whatever lower-tier school(s) happens to be nearby. When you know your competition and what they offer that you do not, it’s very easy to compete: just offer what they offer. For things that aren’t as timely and linked to a knowing how a certain business runs, there is no problem in creating a lifestyle and work ethic that is free-form as long as it meets goals and needs of the firm. (Note: This isn’t my unique idea. A certain investment bank with a strong brand does this sort of thing already… but I did think of it before I knew that!)

Structural Frame 4: Technology, especially experimental or new technologies, should be used to try to create, or even drive, value.

Structural Notes: This is more a philosophy than an actual directive, but it’s important to taking a firm’s strategy on technology to the “next level.” There is a massive body of knowledge within a firm that is lost everyday due to a lack of effort. Usually the solution is to put humans somewhere and have them manually type in numbers or perform mundane tasks to get this working smoothly. Not in our investment bank! Let me furnish you with an example. The corporate bond market works in an unusual way: traders send around “runs” or lists of bonds with quotes of where they are willing to buy and sell bonds via Bloomberg’s messaging system–they are generally free form text. Why do they do it this way? It’s quick and easy. The firm I worked at didn’t make any effort to collect these pricing levels and store them somewhere. However, for publishing strategy reports, helping the desk find trade ideas based on historical relationships, calculating risk metrics, and any other number of things, this data would have been vital. Technology can easily help to store, warehouse, and serve these sorts of datasets (readily available from the market but unstructured) and help the organization as a whole improve its efficiency. This is just one example, but it serves to illustrate a point that is extremely common in an investment bank–lots of things require information that no one keeps but was readily available. Technology can drive value for lots of internal things by helping to solve problems like this. And, honestly, there are too many things that are out of one’s control not to have an organized and structured solution to the simple things that can be fixed.

Another note on technology, however, is that as the Web innovates social behaviors and collaboration those technologies should be actively examined as potential solutions to problems an investment bank would face. For example, lots and lots of information is needed when talking to a client. Getting good market “color” that everyone can see, and that is available, consistent, and easy to find is important. Perhaps a series of blogs could be used to ensure the delivery of this content is made as efficient as possible. One way I added value at my firm was by knowing as many people as possible. When liquidity started becoming an issue, the people I spoke to on the desk that funded banks in the LIBOR-based funding market explained what was the situation and we were able to assess if we thought this warranted a change in our positions or business in general. If that desk had a blog where they posted color throughout the day and the firm had an easy way to deliver this information to all of its employees, perhaps this could have helped people develop a more specific view on the market and notice some irregularities leading to the current crisis. Could Wikis be used effectively? I’m sure that they could. If it was institutionalized to have an up-to-date knowledge base within the firm, and it was made a priority to keep those things updated, nuances and details on complex transactions could be documented. People could avoid falling into the same traps or having to research the same issues other already have. These are just a few examples of how new technology innovations can be used to create value where it would otherwise be impossible.

Structural Frame 5: Every employee should be very comfortable with technology and make a large effort to integrate it into their work.

Structural Notes: I hate to sound like a snob, but in general, if you can’t figure out things like email and basic spreadsheets, you don’t have a lot of room left to grow. People should learn new technologies as they are available and make an effort to work more effectively. If this isn’t a priority of almost everyone in the firm, then building new systems and integrating things into their daily “workflow” is useless. Part of pushing the envelope on how new things are used means that people will have to learn how to use them. I’ve seen too many people, uncomfortable with a new system, resort to keeping their risk positions and other vital data the firm should know in a spreadsheet. Unacceptable. Now, not everyone has to “ooohhh” and “aaahhhh” over new features and technological platforms, but everyone should be asking themselves how they can use some new technology product to make more money, pitch more transactions, better monitor the firm’s risk, develop a better strategy for investing, or whatever their job entails. I don’t think this is hard, but I do think it’s important. And, with technology employees sitting with business people and understanding how they work day-to-day, the resources to figure out these sorts of things will be much more readily available than at most other firms. (See how the “structural frames” all interplay?)

The Final Inspection

As one can see, I value the little things that help people get 10-15% more productivity out of their daily routine–that’s the edge most firm’s need to excel in what they are focusing on. However, most firms poorly thought out systems and infrastructure issues, especially when it comes to technology, adds a hugely cost-ineffective layer of one-fix-at-a-time solutions that have added up. Why have a system where traders can input their own trades as they do them? Give them a paper record and hire a person, with full benefits and being paid an amount commensurate with living in New York City, to type them in. Oh, and now that the business has grown to three times to trading volume in six months, let’s hire four more people. Why have a system that allows a capital markets person to view real-time quotes in their sector or updates those quotes into a spreadsheet or presentation? Just have a bunch of analysts do it by hand. Why would you want a system that can model securitizations and CDOs and run the numbers effectively? We can have someone do it in a spreadsheet, that’s “close enough.” Although it doesn’t capture the nuanced risk factors, I’m sure defaults will never get high enough to worry about. These are the kind of solutions that, from the start, one should be thinking about. From the first instant it’s possible to fix these, they should be fixed. I think the five parts of the framework I’ve laid out will make a good plan to follow when building the technology part of our investment bank!