Books by Lewis, Michael

Lewis, Michael. The Big Short. New York: W. W. Norton, 2010. ISBN 978-0-393-07223-5.

After concluding his brief career on Wall Street in the 1980s, the author wrote Liar's Poker, a memoir of a period of financial euphoria and insanity which he assumed would come crashing down shortly after his timely escape. Who could have imagined that the game would keep on going for two decades more, in the process raising the stakes from mere billions to trillions of dollars, extending its tendrils into financial institutions around the globe, and fuelling real estate and consumption bubbles in which individuals were motivated to lie to obtain money they couldn't pay back to lenders who were defrauded as to the risk they were taking?

Most descriptions of the financial crisis which erupted in 2007 and continues to play out at this writing gloss over the details, referring to “arcanely complex transactions that nobody could understand” or some such. But, in the hands of a master explainer like the author, what happened isn't at all difficult to comprehend. Irresponsible lenders (in some cases motivated by government policy) made mortgage loans to individuals which they could not afford, with an initial “teaser” rate of interest. The only way the borrower could avoid default when the interest rate “reset” to market rates was to refinance the property, paying off the original loan. But since housing prices were rising rapidly, and everybody knew that real estate prices never fall, by that time the house would have appreciated in value, giving the “homeowner” equity in the house which would justify a higher grade mortgage the borrower could afford to pay. Naturally, this flood of money into the housing market accelerated the bubble in housing prices, and encouraged lenders to create ever more innovative loans in the interest of “affordable housing for all”, including interest-only loans, those with variable payments where the borrower could actually increase the principal amount by underpaying, no-money-down loans, and “liar loans” which simply accepted the borrower's claims of income and net worth without verification.

But what financial institution would be crazy enough to undertake the risk of carrying these junk loans on its books? Well, that's where the genius of Wall Street comes in. The originators of these loans, immediately after collecting the loan fee, bundled them up into “mortgage-backed securities” and sold them to other investors. The idea was that by aggregating a large number of loans into a pool, the risk of default, estimated from historical rates of foreclosure, would be spread just as insurance spreads the risk of fire and other damages. Further, the mortgage-backed securities were divided into “tranches”: slices which bore the risk of default in serial order. If you assumed, say, a 5% rate of default on the loans making up the security, the top-level tranche would have little or no risk of default, and the rating agencies concurred, giving it the same AAA rating as U.S. Treasury Bonds. Buyers of the lower-rated tranches, all the way down to the lowest investment grade of BBB, were compensated for the risk they were assuming by higher interest rates on the bonds. In a typical deal, if 15% of the mortgages defaulted, the BBB tranche would be completely wiped out.

Now, you may ask, who would be crazy enough to buy the BBB bottom-tier tranches? This indeed posed a problem to Wall Street bond salesmen (who are universally regarded as the sharpest-toothed sharks in the tank). So, they had the back-office “quants” invent a new kind of financial derivative, the “collateralised debt obligation” (CDO), which bundled up a whole bunch of these BBB tranche bonds into a pool, divided it into tranches, et voilà, the rating agencies would rate the lowest risk tranches of the pool of junk as triple A. How to get rid of the riskiest tranches of the CDO? Lather; rinse; repeat.

Investors worried about the risk of default in these securities could insure against them by purchasing a “credit default swap”, which is simply an insurance contract which pays off if the bond it insures is not repaid in full at maturity. Insurance giant AIG sold tens of billions of these swaps, with premiums ranging from a fraction of a percent on the AAA tranches to on the order of two percent on BBB tranches. As long as the bonds did not default, these premiums were a pure revenue stream for AIG, which went right to the bottom line.

As long as the housing bubble continued to inflate, this created an unlimited supply of AAA rated securities, rated as essentially without risk (historical rates of default on AAA bonds are about one in 100,000), ginned up on Wall Street from the flakiest and shakiest of mortgages. Naturally, this caused a huge flow of funds into the housing market, which kept the bubble expanding ever faster.

Until it popped.

Testifying before a hearing by the U.S. House of Representatives on October 22nd, 2008, Deven Sharma, president of Standard & Poor's, said, “Virtually no one—be they homeowners, financial institutions, rating agencies, regulators, or investors—anticipated what is occurring.” Notwithstanding the claim of culpable clueless clown Sharma, there were a small cadre of insightful investors who saw it all coming, had the audacity to take a position against the consensus of the entire financial establishment—in truth a bet against the Western world's financial system, and the courage to hang in there, against gnawing self-doubt (“Can I really be right and everybody else wrong?”) and skittish investors, to finally cash out on the trade of the century. This book is their story. Now, lots of people knew well in advance that the derivatives-fuelled housing bubble was not going to end well: I have been making jokes about “highly-leveraged financial derivatives” since at least 1996. But it's one thing to see an inevitable train wreck coming and entirely another to figure out approximately when it's going to happen, discover (or invent) the financial instruments with which to speculate upon it, put your own capital and reputation on the line making the bet, persist in the face of an overwhelming consensus that you're not only wrong but crazy, and finally cash out in a chaotic environment where there's a risk your bets won't be paid off due to bankruptcy on the other side (counterparty risk) or government intervention.

As the insightful investors profiled here dug into the details of the fairy castle of mortgage-backed securities, they discovered that it wouldn't even take a decline in housing prices to cause defaults sufficient to wipe out the AAA rated derivatives: a mere stagnation in real estate prices would suffice to render them worthless. And yet even after prices in the markets most affected by the bubble had already levelled off, the rating agencies continued to deem the securities based on their mortgages riskless, and insurance against their default could be bought at nominal cost. And those who bought it made vast fortunes as every other market around the world plummeted.

People who make bets like that tend to be way out on the tail of the human bell curve, and their stories, recounted here, are correspondingly fascinating. This book reads like one of Paul Erdman's financial thrillers, with the difference that the events described are simultaneously much less probable and absolutely factual. If this were a novel and not reportage, I doubt many readers would find the characters plausible.

There are many lessons to be learnt here. The first is that the human animal, and therefore the financial markets in which they interact, frequently mis-estimates and incorrectly prices the risk of outcomes with low probability: Black Swan (January 2009) events, and that investors who foresee them and can structure highly leveraged, long-term bets on them can do very well indeed. Second, Wall Street is just as predatory and ruthless as you've heard it to be: Goldman Sachs was simultaneously peddling mortgage-backed securities to its customers while its own proprietary traders were betting on them becoming worthless, and this is just one of a multitude of examples. Third, never assume that “experts”, however intelligent, highly credentialed, or richly compensated, actually have any idea what they're doing: the rating agencies grading these swampgas securities AAA had never even looked at the bonds from which they were composed, no less estimated the probability that an entire collection of mortgages made at the same time, to borrowers in similar circumstances, in the same bubble markets might all default at the same time.

We're still in the early phases of the Great Deleveraging, in which towers of debt which cannot possibly be repaid are liquidated through default, restructuring, and/or inflation of the currencies in which they are denominated. This book is a masterful and exquisitely entertaining exposition of the first chapter of this drama, and reading it is an excellent preparation for those wishing to ride out, and perhaps even profit from the ongoing tragedy. I have just two words to say to you: sovereign debt.

July 2010 

Lewis, Michael. Flash Boys. New York: W. W. Norton, 2014. ISBN 978-0-393-24466-3.

Back in the bad old days before regulation of financial markets, one of the most common scams perpetrated by stockbrokers against their customers was “front running”. When a customer placed an order to buy a large block of stock, which order would be sufficient to move the market price of the stock higher, the broker would first place a smaller order to buy the same stock for its own account which would be filled without moving the market very much. Then the customer order would be placed, resulting in the market moving higher. The broker would then immediately sell the stock it had bought at the higher market price and pocket the difference. The profit on each individual transaction would be small, but if you add this up over all the volume of a broker's trades it is substantial. (For a sell order, the broker simply inverts the sense of the transactions.) Front running amounts to picking the customer's pocket to line that of the broker: if the customer's order were placed directly, it would execute at a better price had it not been front run. Consequently, front running has long been illegal and market regulators look closely at transaction histories to detect evidence of such criminality.

In the first decade of the 21st century, traders in the U.S. stock market discovered the market was behaving in a distinctly odd fashion. They had been used to seeing the bids (offers to buy) and asks (offers to sell) on their terminals and were accustomed to placing an order and seeing it hit by the offers in the market. But now, when they placed an order, the offers on the other side of the trade would instantly evaporate, only to come back at a price adverse to them. Many people running hundreds of billions of dollars in hedge, mutual, and pension funds had no idea what was going on, but they were certain the markets were rigged against them. Brad Katsuyama, working at the Royal Bank of Canada's Wall Street office, decided to get to the bottom of the mystery, and eventually discovered the financial equivalent of what you see when you lift up a sheet of wet cardboard in your yard. Due to regulations intended to make financial markets more efficient and fair, the monolithic stock exchanges in the U.S. had fractured into dozens of computer-mediated exchanges which traded the same securities. A broker seeking to buy stock on behalf of a customer could route the order to any of these exchanges based upon its own proprietary algorithm, or might match the order with that of another customer within its own “dark pool”, whence the transaction was completely opaque to the outside market.

But there were other players involved. Often co-located in or near the buildings housing the exchanges (most of which are in New Jersey, which has such a sterling reputation for probity) were the servers of “high frequency traders” (HFTs), who placed and cancelled orders in times measured in microseconds. What the HFTs were doing was, in a nutshell, front running. Here's how it works: the HFT places orders of a minimum size (typically 100 shares) for a large number of frequently traded stocks on numerous exchanges. When one of these orders is hit, the HFT immediately blasts in orders to other exchanges, which have not yet reacted to the buy order, and acquires sufficient shares to fill the original order before the price moves higher. This will, in turn, move the market higher and once it does, the original buy order is filled at the higher price. The HFT pockets the difference. A millisecond in advance can, and does, turn into billions of dollars of profit looted from investors. And all of this is not only completely legal, many of the exchanges bend over backward to attract and support HFTs in return for the fees they pay, creating bizarre kinds of orders whose only purpose for existing is to facilitate HFT strategies.

As Brad investigated the secretive world of HFTs, he discovered the curious subculture of Russian programmers who, having spent part of their lives learning how to game the Soviet system, took naturally to discovering how to game the much more lucrative world of Wall Street. Finally, he decides there is a business opportunity in creating an exchange which distinguishes itself from the others by not being crooked. This exchange, IEX, (it was originally to be called “Investors Exchange”, but the founders realised that the obvious Internet domain name, investorsexchange.com, could be infelicitously parsed into three words as well as two), would include technological constraints (including 38 miles of fibre optic cable in a box to create latency between the point of presence where traders could attach and the servers which matched bids and asks) which rendered the strategies of the HFTs impotent and obsolete.

Was it conceivable one could be successful on Wall Street by being honest? Perhaps one had to be a Canadian to entertain such a notion, but in the event, it was. But it wasn't easy. IEX rapidly discovered that Wall Street firms, given orders by customers to be executed on IEX, sent them elsewhere to venues more profitable to the broker. Confidentiality rules prohibited IEX from identifying the miscreants, but nothing prevented them, with the brokers' permission, from identifying those who weren't crooked. This worked quite well.

I'm usually pretty difficult to shock when it comes to the underside of the financial system. For decades, my working assumption is that anything, until proven otherwise, is a scam aimed at picking the pockets of customers, and sadly I have found this presumption correct in a large majority of cases. Still, this book was startling. It's amazing the creepy crawlers you see when you lift up that piece of cardboard, and to anybody with an engineering background the rickety structure and fantastic instability of what are supposed to be the capital markets of the world's leading economy is nothing less than shocking. It is no wonder such a system is prone to “flash crashes” and other excursions. An operating system designer who built such a system would be considered guilty of malfeasance (unless, I suppose, he worked for Microsoft, in which case he'd be a candidate for employee of the year), and yet it is tolerated at the heart of a financial system which, if it collapses, can bring down the world's economy.

Now, one can argue that it isn't such a big thing if somebody shaves a penny or two off the price of a stock you buy or sell. If you're a medium- or long-term investor, that'll make little difference in the results. But what will make your blood boil is that the stock broker with whom you're doing business may be complicit in this, and pocketing part of the take. Many people in the real world look at Wall Street and conclude “The markets are rigged; the banks and brokers are crooked; and the system is stacked against the investor.” As this book demonstrates, they are, for the most part, absolutely right.

May 2014 

Lewis, Michael. Moneyball. New York: W. W. Norton, [2003] 2004. ISBN 0-393-32481-8.

Everybody knows there's no faster or more reliable way to make a lot of money than to identify an inefficiency in a market and arbitrage it. (If you didn't know that, consider it free advice and worth everything you paid for it!) Modern financial markets are Hellishly efficient. Millions of players armed with real-time transaction data, massive computing and database resources for data mining, and more math, physics, and economics Ph.D.s than a dozen Ivy League campuses are continuously looking for the slightest discrepancy between price and value, which more or less guarantees that even when one is discovered, it won't last for more than a moment, and that by the time you hear about it, it'll be long gone. It's much easier to find opportunities in slower moving, less intensely scrutinised fields where conventional wisdom and lack of imagination can blind those in the market to lucrative inefficiencies. For example, in the 1980s generic personal computers and graphics adaptors became comparable in performance to special purpose computer aided design (CAD) workstations ten times or more as costly. This created a situation where the entire value-added in CAD was software, not hardware—all the hardware development, manufacturing, and support costs of the existing vendors were simply an inefficiency which cost their customers dearly. Folks who recognised this inefficiency and moved to exploit the opportunity it created were well rewarded, even while their products were still being ridiculed or ignored by “serious vendors”. Opportunities like this don't come around very often, and there's a lot of luck involved in being in the right place at the right time with the skills and resources at hand to exploit one when you do spot it.

But just imagine what you could do in a field mired in tradition, superstition, ignorance, meaningless numbers, a self-perpetuating old boy network, and gross disparities between spending and performance…Major League Baseball, say? Starting in the 1970s and 80s, Bill James and a slowly growing group of statistically knowledgeable and scientifically minded baseball fanatics—outsiders all—began to look beyond conventional statistics and box scores and study what really determines how many runs a team will score and how many games it will win. Their results turned conventional wisdom completely on its head and that, combined with the clubbiness of professional baseball, caused their work to be utterly ignored until Billy Beane became general manager of the Oakland A's in 1997. Beane and his statistics wizard Paul DePodesta were faced with the challenge of building a winning team with a budget for player salaries right at the bottom of the league—they had less to spend on the entire roster than some teams spent on three or four superstar free agents. I've always been fond of the phrase “management by lack of alternatives”, and that's the situation Beane faced. He took on board the wisdom of the fan statisticians and built upon it, to numerically estimate the value in runs—the ultimate currency of baseball—of individual players, and compare that to the cost of acquiring them. He quickly discovered the market in professional baseball players was grossly inefficient—teams were paying millions for players with statistics which contributed little or nothing to runs scored and games won, while players with the numbers that really mattered were languishing in the minors, available for a song.

The Oakland A's are short for “Athletics”, but under Beane it might as well have been “Arbitrageurs”—trading overvalued stars for cash, draft picks, and undervalued unknowns spotted by the statistical model. Conventional scouting went out the window; the A's roster was full of people who didn't look like baseball players but fit the mathematical profile. Further, Beane changed the way the game was played—if the numbers said stolen bases and sacrifice bunts were a net loss in runs long-term, then the A's didn't do them. The sportswriters and other teams thought it was crazy, but it won ball games: an amazing 103 in 2002 with a total payroll of less than US$42 million. In most other markets or businesses competitors would be tripping over one another to copy the methods which produced such results, but so hidebound and inbred is baseball that so far only two other teams have adopted the Oakland way of winning. Writing on the opening day of the 2004 World Series, is is interesting to observe than one of those two is the Boston Red Sox. I must observe, however, amongst rooting for the scientific method and high fives for budget discipline and number crunching, that the ultimate product of professional baseball is not runs scored, nor games, pennants, or World Series won, but rather entertainment and the revenue it generates from fans, directly or indirectly. One wonders whether this new style of MBAseball run from the front office will ultimately be as enjoyable as the intuitive, risk-taking, seat of the pants game contested from the dugout by a Leo Durocher, Casey Stengel, or Earl Weaver. This superbly written, fascinating book is by the author of the almost indescribably excellent Liar's Poker. The 2004 paperback edition contains an Afterword recounting the “religious war” the original 2003 hardcover ignited. Again, this is a book recommended by an anonymous visitor with the recommendation form—thanks, Joe!

October 2004