Tuesday, November 29, 2005

Yield curve inverts

The 5 yr yield was briefly below the 2 yr yield today. Is the market predicting a recession? I guess historically the Fed always overtightens...

Imagine you want to model interest rate fluctuations. You have to have a decent model for fluctuations at all maturities. Are there any reasonable constraints? How far can the curve invert? How likely is that? What are the self-consistency constraints? How do we incorporate current option prices into the analysis? A well-known practitioner once told me that modeling the yield curve is to modeling equity prices as quantum field theory is to ordinary quantum mechanics!

See here for Derman's explanation of the BDT (Black, Derman, Toy) yield curve model. I discussed a nice bio of Fischer Black here, and Derman's book here.

Ben Stein on globalization

Ben Stein is most likely familiar to you as a comedian in movies and on TV. Earlier in life he attended Yale Law School and worked in the Nixon White House as a speechwriter. Below he gives the positive spin on globalization. However, at the very end he addresses the winners and losers issue, and his description of losers sounds a lot more like the average american than his description of winners!

China: Friend or Foe?
Wednesday, November 23, 2005

A few nights ago, I had the pleasure of speaking to and MC'ing the annual dinner of the Semiconductor Industry Association (SIA). Attending were the major players in the chip world, the Texas Instruments, AMDs, National Semiconductors, and -- well, you get the picture. Very successful, super smart men and women filled the room -- real geniuses who did things like create programs that read and check immense software codes written in India, only do it in real time over the internet. These are people who started and run incredibly complex and productive businesses.

You might have thought it would be a rollicking good time, but it was not at all.

The three main speakers -- George Scalise, head of the SIA; Charlene Barshefsky, formerly Clinton's Special Trade Representative; and Byron Wien, a major power at Morgan Stanley -- were deeply worried about Far Eastern competition in the chip world. It is simple enough: China, Taiwan, Thailand, and India have far lower labor costs than we do. They have up-to-date machinery and a highly trained, well motivated labor force. They can make chips for less, and they are starting to become players in a big way -- not just in chip manufacturing (or "fabbing," as they say) but in design, which was formerly an American fiefdom.

Trees Do Not Grow to the Sky

The three main speakers showed charts and graphs pointing out Asian inroads and the declining U.S. share of production and design. And one speaker expressed doubt that we would see our kids live as well as we do because of Asian competition.

This prompted several thoughts in my wooly head.

First of all, we start out very far ahead of China in wealth and income. The per capita GDP of the U.S., depending on how you measure it, is as much as 40 times Chinese per capita GDP and not less than 12 times by any measurement.

Yes, China is growing much faster than we are, at roughly 10 percent per year. If current trends last, China will overtake the U.S. in per capita GDP sometime this century. But current rates for China are extremely high. No nation has ever been able to sustain the rates China is experiencing for very long. History argues that China will not be able to do so either. Put simply, trees do not grow to the sky.

However, the Chinese are an amazing, intelligent, hard working, well disciplined people. Possibly they will be able to enjoy super high growth rates for a prolonged period. It would be a historical anomaly, but maybe it will happen.

Is What's Good for China Good for the U.S.?

Even if China became much richer than it is now though, that would not necessarily make us poorer. China is a prodigious buyer of foods and resources, and we have a lot of both. Chinese industry will be partly owned by U.S. investors, and Chinese prosperity will (if no fraud is involved) make them much richer. The fact that a very large nation like China becomes rich makes individual American workers who compete with China worse off, but it does not make America as a nation worse off.

Next, is it not interesting that China, in a state of brutal totalitarian rule, was really not a threat to us except in fiction? But under capitalism, China is making us all run around screaming in fear. I am really not sure what we have to fear from China. The country is our partner in prosperity, not our enemy. The earlier we work out a durable framework of respect and cooperation in trade and policy, the better off everyone will be.

But the main thing I want to express, as I did to the SIA, is that China is getting rich because its people are getting well educated, because they are working hard, because they are saving, and because they are investing. These are exactly the same things that made America rich. And these are exactly the same things that are allowing Hispanic and Asian immigrants to America to move up the economic ladder.

Who Wins, Who Loses?

Young Americans who study hard, learn serious subjects, do not lose themselves in computer games, avoid doomed industries, learn good work habits, save prudently, and invest sensibly will be well off no matter what happens in China or Taiwan or India.

Americans who are slothful, do not pay attention to economic trends, learn no useful skills, do not save, and do not invest wisely will roll downhill fast.

But for the disciplined among us who learn from the Chinese the keys to wealth as the Chinese learned from us and we learned from all of history, the future is bright.

Saturday, November 26, 2005

China book reviews

From The NY Review of Books. I briefly mentioned both of these books in earlier posts here and here.

Three Billion New Capitalists: The Great Shift of Wealth and Power to the East
by Clyde Prestowitz
Basic Books, 321 pp., $26.95

China, Inc.: How the Rise of the Next Superpower Challenges America and the World
by Ted C. Fishman
Scribner, 342 pp., $26.00

The "rise" of China has suddenly become the all-absorbing topic for those professionally concerned with the future of the planet. Will the twenty-first century be the Chinese century, and, if so, in what sense? Will China's rise be peaceful or violent? And how will this affect the United States, the current "hyperpower"? In fact, China has been "rising" for some time (after several hundred years of "fall"), but for many years its claim to notice was obscured by more exciting events. Attention in the 1990s concentrated on the fall of Soviet communism, "globalization," the spread of democracy, and the high-tech revolution. These developments, which left America as the world's sole economic and political superpower, seemed to belie Paul Kennedy's prediction in 1987 of relative US decline and "more of a multipolar system."[1]

The attack on the World Trade Center in 2001, together with the concurrent collapse of the high-tech bubble, exposed America's fragility, but this was masked by the hyperactivity of the Bush administration. The "war on terror" planted American armies in Afghanistan and Iraq; the Clinton surpluses were succeeded by the Bush deficits to shore up the economy and finance the military operations. However, as the Iraq escapade foundered and the deficits ballooned, the sense of relative decline reasserted itself. Unlike in 1987, there was now a clear candidate for the succession: China. This was especially so as the US economy became dependent on China's bankrolling its huge trade deficit. The dream of an "American century" receded, to be replaced by the nightmare of a "Chinese century."

Focus on China is overdue. For the last quarter of a century its economy has been growing by over 9 percent a year, increasing eightfold. However, it is not just this long-sustained hyper-growth rate that amazes and alarms the observer. It is the size of the economy which is growing. China's population is officially estimated at 1.3 billion, but is probably larger—one fifth of all the people in the world. This makes its rise much more important than that, say, of Japan in the 1960s. From the economic point of view its cheap labor is much more abundant, so its cost advantage will not quickly be eliminated. The size of an economy obviously matters, too, in measuring power. The Chinese economy, in terms of the purchasing power of the Chinese people, is about two thirds the size of the US economy.[2] If it continues to grow at 9 percent a year, it will overtake the US by 2014. Lee Kwan Yu of Singapore believes that the rise of China will shift the balance of power back to the East for the first time since Portuguese caravels arrived there in the sixteenth century.

China's growth, simply because of its size, is bound to create problems both for itself and others. From the Chinese leadership's point of view, the main problem is how to maintain social cohesion amid the vast socio-economic upheavals going on. Apart from the environmental degradation and rampant corruption, China's pell-mell, and largely uncontrolled, economic growth is disturbing its domestic stability in a profound way: there is a huge floating population without settled jobs or abodes, and a development and income gap between the coastal and inland areas which is as big as between the United States and North Africa. According to one estimate, 30 percent of China's urban workforce, or 200 million people, is currently unemployed or underemployed. The livelihood of another 100 million agricultural workers is threatened as World Trade Organization rules increase China's dependence on foreign food supplies. The specter of chaos frightens the rulers in Beijing.

In international relations, the issue is whether China' impact on the world will be peaceful or violent. Th debate here follows disciplinary lines. "Those who focus on economics tend to see partnership, cooperation and reasons for optimism despite tensions, while security experts are more pessimistic and anticipate strategic conflict as the likely future for two political systems that are so different," writes one commentator [3]. Both views can claim some evidence in their favor...

Friday, November 25, 2005

Google click to call

This is the next step in Internet advertising. Among the text ads that come with your search results you might see a little telephone icon. If you click it and enter your phone number, Google will connect a representative from the advertiser to you via VOIP. (Your phone will ring, and voila, you can discuss how the refrigerator gets delivered or whether the plasma screen is glare resistant. Guess what part of the world the call center rep will be working from :-) This sort of thing is already out there using IM-type clients, but here the user only needs an ordinary phone. Google does not tell the advertiser your phone number, so your anonymity is protected. Very nice!

I'm very envious of the Google people because they actually have the opportunity to implement all of these nifty ideas.

Sunday, November 20, 2005

Simons, Thorp and Shannon

Nice article on Jim Simons and Renaissance in Saturday's Times. Part of me wants to get into the new fund (fees are pretty reasonable compared to Medallion), but then again they are embarking on something new, so it's no sure thing. See earlier posts here and here. Most impressive about Medallion is their consistency -- no down years since 1988 and only a single down month in the last 5 years!

BTW, I've received my copy of Fortune's Formula and it's quite good. I learned a number of tidbits about Thorp (the mathematician who wrote Beat the Dealer and invented a system for counting cards in blackjack), Shannon (the father of information theory) and others from this book. Apparently, Thorp and Shannon's investment returns (Thorp ran an early hedge fund called Princeton-Newport, while Shannon invested his own account) rivalled those of the best managers like Buffet and Soros. Buffet and Thorp actually knew each other early on, and had very high opinions of each other. The stories of Thorp testing his card counting system in Nevada are hilarious -- the level of detail after all these years suggests a phenomenal memory!

The bit about optimizing geometric vs arithmetic returns (a subject of controversy between math/physics guys like Kelly, Shannon, Thorp and economists such as Samuelson, and the origin of the title of the book) seems not so interesting to me, as the answer depends on what one wants to achieve. On the subject of hedge funds, it appears everyone is starting one, including information theorist Thomas Cover and former physicist turned AI researcher Eric Baum (author of What is Thought?, the best book I've read on AI).

NYTimes: $100 Billion in the Hands of a Computer

PEOPLE ask me all the time: What's your secret?" James Simons said. We were sitting in an office in Manhattan that Mr. Simons uses when he's not at the Long Island offices of Renaissance Technologies, the money management firm he founded in 1982. He was wearing an elegant shirt and tie, and loafers with no socks. He took a drag from a cigarette, the second of three he would smoke in the course of a long interview.

I had indeed come to ask him what his secret was. In the hedge fund world, that's what everybody wants to know.

Mr. Simons, 67, who rarely talks to journalists, is hardly a household name like Warren E. Buffett. But Mr. Simons, who got into the hedge fund business after abandoning a stellar career in mathematics, has a track record that is jaw-dropping. This summer, word leaked out that he was starting a new fund - people took to calling it the "$100 billion fund" because its marketing materials say that it could conceivably grow to that enormous size. Not surprisingly, that has caused Wall Street types to be even more curious about him.

Here are Mr. Simons's numbers: from 1990 to 2004, Renaissance's primary hedge fund, called Medallion, has delivered annualized returns of 33.21 percent. (The Standard & Poor's 500-stock index has returned, on average, 10.98 percent during those same years.) Since the end of 2002, the fund, which has $5 billion under management, has disbursed $4.9 billion to its investors - with another $1.5 billion to be delivered at the end of this year.

And these returns are after Medallion's 5 percent management fee and 44 percent share of the profits - surely the highest hedge fund fees in the land. Medallion's returns, and its fees, have helped make Mr. Simons a very wealthy man, with a net worth that Forbes estimates at $2.7 billion.

When I showed Mr. Simons's returns to a hedge fund friend, he looked startled. "Nobody has numbers like those," he said. But here's the real eye-opener: no one outside the firm's 200 or so employees has a clue how he does it.

Medallion, you see, is a quantitative fund. In quant funds, trading activity is generated by complex computer models rather than human judgment. Most quants are secretive about the algorithms that drive their models; after all, that's their investing edge. But of the handful of big-time "black box" investors, as they're often called, Mr. Simons's box may well be the blackest.

HERE'S what we do know. Medallion's portfolio contains literally thousands of stocks and other financial instruments that it trades in rapid-fire fashion. The firm's scientists are constantly searching for repeatable patterns, and other signals, in the enormous amounts of data they compile. The computer models they devise tell them when to make trades based on those signals.

As Mr. Simons put it - and this is about as specific as he would get - "Certain price patterns are nonrandom and will lead to a predictive effect." He also told me that Medallion sticks with highly liquid securities that trade in public markets around the world. Why? "Because there is a lot of data on such instruments, and we're very statistically oriented," he said. He stays away from exotic derivatives.

Not even Mr. Simons's investors know much more than I've just described. "We trust Jim and we think he's smart," said one longtime Medallion investor. "So we stopped caring what the computer was doing." When this investor began describing Mr. Simons's investing approach, he admitted he was guessing.

Mr. Simons shrugged when I suggested to him that his firm's lack of "transparency," as they say in the business, was bound to make people nervous. Humans fail in the market all the time, but somehow we are willing to keep giving our money to human beings to manage because we understand investing based on human judgment. Or at least we think we do. But black box investing feels different. It feels scary somehow, precisely because it is not something most of us can understand.

"How any great investor does it isn't in the least obvious," Mr. Simons responded. "How we do it isn't any more mysterious than how a great fundamental investor does it. In some ways it is less mysterious because what we do can be programmed." Then he stopped, took another drag from his cigarette, and let out a small chuckle. "Well," he conceded, "it's less mysterious to us."

Mr. Simons wasn't always a quant. A former crypt analyst - a code breaker, that is - he did important work in mathematics that helped lay the foundation for string theory. When he began managing money in the 1970's, he did it the same way most investors did: he used his own judgment. "At first," he said, "I didn't think about investing in a scientific fashion. But I was trading currencies, and it gradually occurred to me that there might be some way to create models that would allow you to predict currency movements."

Although Mr. Simons and a partner made an absolute killing in the currency markets the old-fashioned way - they made huge bets that turned out to be right - he began surrounding himself with scientists who developed models for all sorts of tradeable securities. "By the end of the 1980's," he said, "I was a model man, and didn't want to do fundamental analysis." One advantage, he said, is that "models can lower your risk." Another, though, is that "it reduces the daily aggravation." With old-fashioned stock picking, he said: "One day you feel like a hero. The next day you feel like a goat. Either way, most of the time it's just luck."

Indeed, trading the way he does, making thousands of small trades aimed at capturing small price movements, doesn't generate the kind of "10 bagger" that investors love. But when done well, quant investing is less likely to have the kind of disaster that is always the danger when one bets big on a stock.

To those who point to Long-Term Capital Management as an example of the dangers of black box investing, Mr. Simons's defenders point out that his fund has far less leverage than Long-Term Capital, and that in any case, while Long-Term Capital had several Nobel laureates on board, human bets were what caused it to go awry.

Clifford Asness, another well-known quant hedge fund manager, said that while he knew no more about Mr. Simons's methods than anyone else, "It's hard to believe that there isn't a measure of safety in Jim's approach.

"Presumably, he's got a highly diversified portfolio, high turnover, and he's capturing small inefficiencies. It's hard to lose a ton of money doing that. It is always possible that someday his models might stop working. But that's different from 'blowing up.' "

"You know," Mr. Asness added, "human beings have a black box, too. It's called the brain."
As for the new "$100 billion fund," Mr. Simons was even more constrained than usual, thanks to regulatory restrictions that limit what he can say publicly while the fund is raising money. People are buzzing about it nonetheless, for it seems to be a major departure from Medallion. Medallion's investors were almost all wealthy individuals; the new fund, called the Renaissance Institutional Equities Fund, has a $20 million minimum investment and is aimed at institutions. It has a much lower fee structure. It will invest in - or sell short - only publicly traded equities. Instead of making rapid-fire trades, it will be much closer to a buy-and-hold portfolio. And so on.

In one critical way, though, it is similar to Medallion. As the marketing document, which I obtained from a person unconnected to Mr. Simons, put it: "The company's risk control, variance and covariance estimation, execution techniques, slippage models, and predictive signals are all derived from those employed by the managing member in trading the Medallion Funds."

In other words, Mr. Simons believes that computer models similar to those that have worked for Medallion will also work for a fund that can hold $100 billion worth of stocks over long periods of time. It is absolutely audacious.

What interested me most of all was: why? At an age when most men are contemplating retirement, with more money than he can count, why was Mr. Simons still at it? "I enjoy the challenge," he replied.

He then began describing a demonstration he saw recently of a new nuclear accelerator at the Brookhaven National Laboratory, where he is on the board. Two atoms hurtled toward each other, colliding with great force. "A huge number of particles are thrown out," he said, "and the job is to analyze everything that results from the collision."

"Watching the spray of particles on the screen made me think of the stock market," he continued. Every trade, even of a hundred shares of a company, affects every other trade. And every day there are thousands upon thousands of such trades, all of them affecting the rest of the market. His work, as he sees it, is to analyze that incredibly complex mosaic and try to figure out how it all fits together.

"The subject may not be the most important in the world," he concluded, "but the dynamics of the market are really interesting. It's a serious question."

I suddenly understood the motivation behind Mr. Simons's new fund. He's doing it because he wants to see if it can be done. Once a scientist, always a scientist.

Saturday, November 19, 2005

The new white flight

Instead of white students fleeing academically weak school districts dominated by disadvantaged blacks and hispanics, in Silicon Valley white students are fleeing overly strong school districts dominated by Asians. Who needs all that math and science anyway?

See related post on Asians and affirmative action here.
WSJ: CUPERTINO, Calif. -- By most measures, Monta Vista High here and Lynbrook High, in nearby San Jose, are among the nation's top public high schools. Both boast stellar test scores, an array of advanced-placement classes and a track record of sending graduates from the affluent suburbs of Silicon Valley to prestigious colleges.

But locally, they're also known for something else: white flight. Over the past 10 years, the proportion of white students at Lynbrook has fallen by nearly half, to 25% of the student body. At Monta Vista, white students make up less than one-third of the population, down from 45% -- this in a town that's half white. Some white Cupertino parents are instead sending their children to private schools or moving them to other, whiter public schools. More commonly, young white families in Silicon Valley say they are avoiding Cupertino altogether.

Whites aren't quitting the schools because the schools are failing academically. Quite the contrary: Many white parents say they're leaving because the schools are too academically driven and too narrowly invested in subjects such as math and science at the expense of liberal arts and extracurriculars like sports and other personal interests.

...In the 1960s, the term "white flight" emerged to describe the rapid exodus of whites from big cities into the suburbs, a process that often resulted in the economic degradation of the remaining community. Back then, the phenomenon was mostly believed to be sparked by the growth in the population of African-Americans, and to a lesser degree Hispanics, in some major cities.

But this modern incarnation is different. Across the country, Asian-Americans have by and large been successful and accepted into middle- and upper-class communities. Silicon Valley has kept Cupertino's economy stable, and the town is almost indistinguishable from many of the suburbs around it. The shrinking number of white students hasn't hurt the academic standards of Cupertino's schools -- in fact the opposite is true.

...white students represented 20% of [Monta Vista's] 29 National Merit Semifinalists this year.

...At Cupertino's top schools, administrators, parents and students say white students end up in the stereotyped role often applied to other minority groups: the underachievers. In one 9th-grade algebra class, Lynbrook's lowest-level math class, the students are an eclectic mix of whites, Asians and other racial and ethnic groups.

"Take a good look," whispered Steve Rowley, superintendent of the Fremont Union High School District, which covers the city of Cupertino as well as portions of other neighboring cities. "This doesn't look like the other classes we're going to."

On the second floor, in advanced-placement chemistry, only a couple of the 32 students are white and the rest are Asian. Some white parents, and even some students, say they suspect teachers don't take white kids as seriously as Asians.

Thursday, November 17, 2005

Outsourcing at Conexant

The handwriting is on the wall...

NYT: Dwight Decker, Conexant's chief executive, said that half the semiconductor design and other high-tech engineering work is now done at Conexant India, the division in Hyderabad.

The operation there employs 700 engineers, nearly as many as headquarters does. That is up from 10 percent of such work last year, and Mr. Decker says that figure will jump to 65 percent by the end of 2006, leaving just one-third of the work to be done by the engineers in the United States.

"We are placing a very large bet on our ability to shift a significant part of our development to Asia," Mr. Decker said. "We are doing that more aggressively than any semiconductor company." He emphasizes that no layoffs are planned for Newport Beach, where engineers will work on the "innovation and architecture" of the firm's semiconductor systems.

Conexant's step is enormous, proportionate to its middling size. Giants like General Electric and 3M routinely assign high-tech research to laboratories they own in India and China. And the sprinkling of such work among their global systems is increasing. But most of their research remains in the United States.

Conexant is a fraction of their size, with only 2,400 employees and $900 million in sales of computer and communications equipment last year. Yet it is a technological powerhouse - the inventor 50 years ago of the computer modem - and a major investor in research and development.

Conexant will spend about $250 million this year on research and development because it must keep coming up with new microelectronic wonders for demanding customers like Samsung Electronics, DirecTV and others that are striving to bring movies, music, medical diagnoses and every conceivable service via Internet television to the digital home. Conexant is a strong and rising competitor in the electronics that make possible broadband Internet reception through D.S.L. telephone lines.

In 1955, as a division of North American Aviation, Conexant developed the transistor modem for communicating data over telephone lines, under a contract from the Defense Department. In the early 1980's, as part of Rockwell International, it brought out high-speed fax modems, and in 1996, it introduced high-speed Internet connectivity.

"We have adapted time and again," said Mr. Decker, who has led Conexant since it was spun off from Rockwell in 1999. "And we will continue to adapt and play our part in an expanding world market as long as we innovate." Conexant is putting emphasis on India to keep ahead of it main competitors, the Broadcom Company and the Swiss-based giant, ST Microelectronics Group, said Mr. Decker, a physicist and mathematician with degrees from McGill University in Montreal and a doctorate in math from the California Institute of Technology.

Mr. Decker said that besides financial reasons the export of research jobs is motivated by a looming dearth of engineering talent in the United States. Engineers in India earn one fourth of the pay of their American counterparts, roughly $25,000 a year in salary and benefits, compared with $100,000.

"If we can get two-thirds of our product development at one-fourth the cost, we come close to cutting our overall costs in half," Mr. Decker said. In the first year of large-scale work in India, he said, Conexant reduced costs by $36 million.

Conexant may be a harbinger of developments at corporate research departments across America. "There is a lot more research work being done in India these days," said Shivbir Grewal, a lawyer in Irvine, Calif., who works with companies in America and India.

The attraction is "the huge pool of talent" from the Indian Institutes of Technology - seven major institutions established over the last 54 years - and regional engineering colleges, Mr. Grewal said. But "visas to the United States have been extremely restricted since 9/11," Mr. Grewal said, "so the graduates are staying home and finding work in India." The increase in research overseas is arousing concerns in the United States. The fear is that good jobs will migrate to India and China and that the innovative wellspring of new technology will slowly dry up in the United States.

Terry Opdendyk, a venture capitalist, disagrees with the second concern, but concedes the first.

"It is extremely difficult to achieve and manage innovation in collaborations across oceans and time zones," said Mr. Opdendyk, who has backed more than 100 start-ups as head of Onset Ventures, a Silicon Valley firm he founded in 1984. "But I worry because we're educating too few engineers in America and I don't see us pursuing change-the-world ideas as we used to," Mr. Opdendyk said.

Mr. Decker agrees that at fewer than 60,000 new engineers a year, the United States may not have enough skilled people to handle all the work to be done. However, he said, the United States still leads in information technology, and "we can sustain our world leadership if we continue to innovate."

Tuesday, November 15, 2005

Fortune's formula

Fortune's Formula: The Untold Story of the Scientific Betting System that Beat the Casinos and Wall Street, by William Poundstone

I just ordered this book from Amazon -- apparently it is in vogue among some of the big institutional investors :-)

The optimization problem described is interesting from an academic point of view, and also as a good rough guide for investors, but of course it assumes that certain quantities, such as correct probabilities of future outcomes, are knowable. That may be the case in casino gambling, but in investing one can only make rough guesses based on past performance. I recall being puzzled at a section on portfolio optimization in the information theory book by Cover and Thomas, but now I see the connection.

BTW, on the topic of finance books, I highly recommend this biography of Fischer Black, which I should have reviewed here long ago. Fischer was yet another outsider (his background was in theoretical physics) to finance who made an important contribution. Unlike Kelly, he was accorded mainstream recognition (professorship at Chicago and partnership at Goldman) during his career. The most impressive thing about Black was his ability to think deeply and independently -- beyond the conventional wisdom. There are some very intriguing passages in the book about his views on money and banking which are, I think, quite unconventional to mainstream economics.

See here for an interesting review of Fortune's Formula by Berkeley math professor E. Berlekamp -- himself a former manager of Jim Simon's Medallion Fund and a collaborator of Claude Shannon!

In a paper published in 1956, John L. Kelly of Bell Labs formulated the asset-allocation problem in terms of an idealized model for which he derived some quantitative results. He used colorful racetrack terminology reminiscent of the classic Damon Runyon movie Guys and Dolls: Suppose that one goes to the racetrack with an available bankroll, B. Suppose further that one knows for each horse the correct probability that it will win the next race. Suppose further that the betting odds are at least slightly inconsistent with this information. And finally, suppose that each race is merely one of a very long sequence of betting opportunities. Kelly found criteria for deciding how much one should then bet on each horse in each race.

Kelly observed that, under similar idealized assumptions, the same formulation could also be applied to investments. In the idealized model, the portfolio manager has an accurate probability distribution on the future performance of each asset in the universe of potential investments. Kelly's methodology then provides a quantitative specification of how big a position to take in each of the candidate assets. Not surprisingly, the fraction of one's portfolio to be invested in any asset that has a negative expected rate of return will be zero. Most assets with positive expected rates of return will merit the investment of some positive fraction of the portfolio. Among assets with similar expected rates of return, those whose returns are relatively stable will be weighted more heavily than those whose future returns have significant risks of substantial losses, even when these risky investments also have some chance of large gains. All of these qualitative features of Kelly's performance criteria concur with conventional wisdom. What distinguishes Kelly's work from that of his predecessors is his quantitative specificity and the fact that he succeeded in proving that, under his assumptions, in the very long run the bankroll of an investor who followed his criteria would eventually surpass the bankroll of anyone following any other strategy.

Kelly also derived a formula for the rate at which this bankroll would grow. This formula is related to a fundamental information-theoretic notion that Claude Shannon (now widely considered to be the father of the information age) had introduced in 1948. Shannon had shown that noise on a communication channel need not impose any bound on the reliability with which information can be communicated across it, because the probability of transmitting a very long file inaccurately can be made arbitrarily small by using sufficiently sophisticated coding techniques, subject to a constraint that the ratio of the length of the source file to the length of the encoded file must be less than a number called the channel capacity. Kelly showed that the asymptotically optimum asset allocation could be determined by solving a system of equations that maximized the log of one's capital. In his horse-track jargon, Kelly also showed that the resulting optimal compound growth rate could be viewed as the capacity of a hypothetical noisy channel over which the bettor was getting the information that distinguished his odds from those of the track. Kelly's betting system, expressed mathematically, is known as the Kelly criterion.

The title of Kelly's paper, "A New Interpretation of the Information Rate," highlighted his discovery of a situation in which Shannon's celebrated capacity theorem applied even though no coding was contemplated. The paper, which appeared in the Bell System Technical Journal, initially attracted a modest audience among information theorists but went unnoticed by economists and professors of finance courses in business schools. Perhaps it would have received more attention if it had had another title. "Information Theory and Gambling" was the title that Kelly himself used for an earlier draft of his paper, but that title was rejected by AT&T executives.

Friday, November 11, 2005

Israeli startups

The Economist discusses the disproportionate impact of Israeli startups in the tech industry. Israelis are especially prominent in computer security, often due to their training in the military. One of the public companies that almost acquired my last startup (since merged/acquired by Juniper) had an extremely bright Israeli CTO who had come aboard from a previous acquisition. Like many other Israeli technologists, he was an alumnus of the IDF (Israel Defense Forces) signal corps training program, which takes in the brightest recruits for intensive IT training. Membership in that unit is sort of equivalent to having a Caltech or MIT degree. At least, the screening process is pretty tough. Of course, there might be other reasons for Israeli success in high tech :-)

Red Herring: Talpiot is a special army training program that puts the best high school graduates through a rigorous curriculum of computer science, physics, and math, then places them in key assignments in, say, intelligence units.

...The selection process for Israel's army-trained technology elite starts when teenagers apply to programs, usually in their last two years of high school. Only volunteers are eligible to be chosen for the army's training programs. The most selective program, Talpiot, accepts only 30 applicants, or 1 in 10, a year. Officers say the army doesn't look for fuzzy traits like creativity and leadership; it focuses on measurable qualities. Extremely high aptitude in math and science, along with success in rigorous exams, are the key qualifications.


Talpiot's M.O. is total immersion, whether the subject is software coding or the Arabic language. The programming course is just six months long, but classes run from 8 a.m. to 10 p.m., five and a half days a week. Although many soldiers say that the program's immersion approach is an effective way to learn, Col. Tregar says it's simply the only way to cram a lot of information into a very short period of time. There's little time to spend on theory and skills that won't directly relate to the students' army postings later on. "In an academic setting, you'll learn about models for parallel processors and the structure of compilers -- those kinds of things," Col. Tregar says. "It's true it gives you a much broader understanding, but on a practical level, you won't have to deal with them in your first job. People with academic degrees that come to the IDF need to undergo considerable training before being put into practical assignments."

Economist: In 2003, 55% of Israel's exports were high technology, compared with the OECD average of 26%. Tech giants such as IBM, Motorola and Cisco have research centres in Israel, which is also where Intel developed its Centrino chip. Not bad for a country with a population of 6.9m.

Why is Israel—sometimes called the “second Silicon Valley”—so strong in technology? For several reasons, says Mr Mlavsky. First, the pump was primed by government grants in the 1970s, by the BIRD Foundation (a joint American-Israeli initiative that supported many start-ups before VC money was widely available), and by government schemes to encourage Russian immigrants who arrived after the collapse of the Soviet Union.

The second big factor is the army. “The army gets hold of everybody at age 18, and if they have a glimmer of potential, it catalyses their transformation into engineers or scientists,” says Mr Mlavsky. The technically minded are given projects to develop and run, and are allowed to keep any intellectual property that they develop, which results in many spin-outs. It also means that once they get to university, trainee engineers already have practical experience and a problem-solving mentality. Israel has 135 engineers per 10,000 employees, compared with 70 in America, 65 in Japan, and 28 in Britain (see chart).

The small size of Israel's home market is also, paradoxically, an advantage. While a British start-up, say, will look to its home market to get started, Israeli firms cannot. Accordingly, they look to America for customers, so that Israeli start-ups function as “mini-multinationals” from the off—and are instantly exposed to the world's most competitive high-tech market. Similarly, Israel's relative lack of land and resources serves to steer entrepreneurs towards high technology instead.

Naturally, cultural factors play a part too. Around 5% of start-ups in America are headed by repeat entrepreneurs, says Mr Mlavsky, compared with around 30% in Israel. “The whole culture, we're like junkies, and the real kick is success, not the fruits of success, so we want to do it again,” he argues. Israeli entrepreneurs are often workaholics who tend not to change their lifestyles much after becoming successful, he says. Gil Shwed, the boss of Check Point and one of Israel's richest men, still has a regular DJ slot at a Tel Aviv restaurant on Wednesday nights, for example.

The bad news for other countries that wish to encourage the development of their technology industries is that few of these factors can be replicated. Singapore's attempt to establish itself as a biotechnology centre faces the challenge of encouraging risk-taking and entrepreneurialism in a highly conformist society. And Britain is hardly likely to introduce conscription in order to boost the fortunes of the technology cluster around Cambridge University. In technology, as in so many other ways, Israel is a special case.

Thursday, November 10, 2005

Chalabi, Hitchens tragicomedy

I don't know whether to laugh or cry. As you may know, Chalabi is the former Iraqi exile (and Chicago math PhD) who conned the neocons with all kinds of crappy pre-war WMD "intelligence," as well as the notion that our troops would be welcomed with flowers (who would you have believed on this, Chalabi or Army Chief of Staff Shinseki?). Rather than being the subject of an FBI/CIA investigation, he is being feted in Washington by Condi and others, and giving a lecture at AEI, apparently attended by Iraq war apologist and self-important blowhard Christopher Hitchens. The little nugget below comes via TalkingPointsMemo.com (see here for more).

Hitchens then turned the subject back to Chalabi, his good friend. I asked him if he thought Chalabi had been passing American intelligence to the Iranians. "No," he insisted. "It's possible that with his training, you know, at [The University of] Chicago that with his own ability he was able to crack the codes. He is a mathematical genius. His expertise is cryptology. It is possible that he broke the codes himself." (This is a paraphrase since I was walking down M Street and crossing Connecticut Avenue all while being amazed that I was having an actual conversation with Christopher Hitchens at the time). Now, I don't believe this for one second. Why would Chalabi be trying to break American codes in his spare time anyway? Who does that if they are friendly to us? Suspicious, I say.

Well, Chalabi's expertise is not cryptography, and no, he didn't single handedly break any Iranian ciphers. What he did was pass on some important intel to the Iranians (that the US had broken their ciphers) that should occasion yet another investigation into neocons leaking classified information.

What makes me most sad is that someone (Hitchens) so ignorant and un-careful in thinking and speaking about topics about which they know nothing could be a public intellectual. But then again in our culture knowledge of basic mathematics or physics is considered geek esoterica whereas an educated person is expected to have read all of Shakespeare.

Monday, November 07, 2005

Baby boom

Invaders: 2, Adults: 0.

The invaders have overwhelmed us! Communications may be interrupted for some time...

Friday, November 04, 2005

VCs, Greenspan and globalization

NYTimes on Silicon Valley VCs and China investments. Note the final paragraph.

WHEN Joe Schoendorf, a Silicon Valley venture capitalist, was in Shanghai a few years ago to hear a pitch from a Chinese start-up company, he sensed something familiar. He interrupted the meeting, walked to the window and pulled back the curtains.

"What are you looking for?" he remembers the would-be entrepreneurs asking.

"I just wanted to make sure I was in China and not back in Palo Alto," he responded.

China's high-technology community, with its brains and competitive spirit, is probably more like its counterpart in Silicon Valley than any other in the world.

Yet Silicon Valley's views of investment in China have tended to swing between wild optimism and deep anxiety - with the anxiety going beyond a fear of losing money. Some worry about helping Chinese start-ups move up the technology food chain.

These days, the Valley venture capitalists are sharply divided in two camps: one rushing into China and one holding back.

...The dominant perspective is that China is a vast sea of opportunity, from its low-cost skilled labor pool to its enormous consumer market that is more than one billion strong.

In fact, it is now routine for venture investors to demand that their start-up firms place the bulk of software development and manufacturing efforts in China or India. (A supply chain problem at a manufacturing arm in China, however, can easily ruin financial results in any given quarter.)

For China skeptics, the concern is that American investment will help energize a formidable competitor, which could come to dominate both markets and technologies.

The fear is based in the Valley's complex relationship with China as supplier, partner, customer and competitor. Most venture capitalists say this evolving relationship will define the future of the Valley and maybe even technology development in the United States.

The Ningbo Bird Company is one case in point. It went from being a contract manufacturing supplier for Motorola to being a serious rival in the Chinese handset market in a matter of a few years.

Still, last year, most of the Valley seemed to throw caution aside as venture firms invested nearly $1.3 billion in China, up nearly 30 percent from 2003, according to Zero2IPO, a venture capital research and consulting company based in Beijing.

But in the first half of this year, investment slowed drastically after several changes in Chinese securities regulations. Those new rules caused "a decline of 50 percent in the first two quarters," said Dixon Doll, managing director of Doll Capital Management, based in Menlo Park, Calif.

The lull is ending, though, in part because of the high-profile success of the initial public offering of Baidu, a Chinese search engine company that was able to raise $86.6 million in August, and a securities rule change in October. In September, Sequoia Capital, a major backer of Google, was reported to be planning a $200 million fund and hiring several employees in China.

That announcement followed an earlier joint agreement this summer by Accel Partners, a leading Silicon Valley firm, and the International Data Group to set up a $250 million fund.

There have even been reports recently that Kleiner Perkins Caufield & Byers, the Valley's highest-profile venture firm, was creating its own China fund, though people briefed on the firm's plans said that was not true. While Kleiner has recently added Colin L. Powell as a partner to serve as a "rainmaker" in Asia, it remains concerned about changes in Chinese security laws that could complicate the return of investment funds to the United States.

Mr. Schoendorf, who is an Accel partner, sees benefits in helping China to become a fierce new competitor. He likens this moment of anxiety and promise to the 1970's, when Japan began to compete successfully with the United States.

"The Chinese graduate more engineers than we do," he said. "They're smart, they work hard, and so the only way to compete with them is to remain more innovative."

WSJ: Greenspan comments on globalization of labor markets. (Holding down inflation is the same as diminished returns to labor...) The bond market reacted negatively to these comments, as Greenspan said the effect would eventually go away. But I think that was a mistake: only the leading edge of those 2 billion workers have been integrated into the world economy -- a distinct minority. It will take decades before that deflationary effect wanes. Not to say that there aren't other inflationary forces at play, like oil and commodity scarcity, but the labor component is deflationary as far as the eye can see. Not only do low-skill workers have little pricing power, but even engineers and service professionals are under pressure from abroad.

The integration of China, India and the former Soviet bloc into the world trading system is helping to hold down inflation but at some point, that effect will fade, Federal Reserve Chairman Alan Greenspan said.

...Mr. Greenspan said the addition of more than 100 million educated workers from former Soviet countries, large segments of China's 750-million strong work force, and workers from India "would approximately double the overall supply of labor once all these workers become fully engaged in competitive world markets," a development that "has restrained the rise of unit labor costs in much of the world and hence has helped to contain inflation."

But while these forces "may well persist for some time," he said, the process and its contribution to inflation control will wane and that "will need to be monitored carefully by the world's central banks."

Thursday, November 03, 2005

Black holes for financiers

No, that's not a typo: it's black holes, not Black Scholes :-) I posted the following comment on Brad Setser's blog (mostly read by finance and econ types). He was fascinated by this Times article on a measurement of the size of the supermassive black hole at the center of our galaxy.

It's now widely believed that many galaxies have at their centers supermassive black holes with masses larger than a million solar masses. (The existence of these supermassive holes is a plausible consequence of how galaxies are formed through non-linear growth of density perturbations in the early universe, but that is another story.) Because black holes are dense, even these supermassive black holes are not that large. The one at the center of our galaxy appears to have diameter roughly that of the earth-sun distance (= 93 million miles; the galaxy is 10^9 times larger), and a mass of 4 million solar masses.

It is easy for astronomers to find objects which are so dense that, theoretically, they must be black holes. The thing at the center of our galaxy is one example; there are many more that are remnants of stellar evolution (very large stars that run out of fuel and collapse to form black holes -- these are only a few orders of magnitude more massive than our sun). To estimate the density astronomers need to measure the mass (based on the gravitational pull exerted on nearby objects such as stars) and size of the object.

The recently reported findings have to do with radio astronomers looking at the black spot caused by absorption of radio waves coming from behind the black hole. They can measure the radius of the black spot to estimate the size of the black hole.

What is most interesting about black holes is the general relativistic effects associated with their "horizon" -- particles, even photons, which pass the horizon can never escape and are doomed to eventually fall into the "singularity" inside the black hole where (according to general relativity) known physics must break down. Unfortunately, based on what I just said, outside observers will never know what happens at the singularity since the object falling in cannot send a signal back out once it has passed the horizon!

To establish the properties of the horizon predicted by general relativity would require probes which signal back to us as they fall in. (They would go dark just as they hit the horizon -- their signal would be "infinitely redshifted.") Astronomical observations are not really enough (although these radio observations are a step in the right direction) since in principle a very dense object which isn't necessarily a black hole might absorb photons (radio waves) to yield a black spot. The identification of an object as a black hole simply based on its density relies on general relativity in a domain in which the theory has not been tested experimentally.

Tuesday, November 01, 2005

Economist on intellectual property

The Economist has a nice survey on patents and IP. As I mentioned in an earlier post, the US patent system (esp. for software) is pretty broken right now. The Economist looks forward to a day when ideas themselves can be bought and sold like any other product. In such a world, some firms might specialize solely in innovation, leaving other tasks like manufacturing, marketing and distribution to others. Of course, this assumes a smoothly functioning IP system, which seems very far off. At the moment, the legal costs of resolving a patent dispute are sufficiently high that startups (which produce a disproportionate amount of innovation) often cannot afford any confrontation with a public company. Given that many big companies are aggressively pursuing a patent-hoarding strategy (hiring more patent attorneys, but perhaps not more scientists!), this may have a negative effect on innovation. See here for leading patent recipients in 2004.

It can take years of hard work, and millions of dollars, for VCs, inventors and entrepreneurs (all of whom have "skin in the game") to determine the worth of an idea. (In this process, the original idea is almost always revised in important ways.) Thus, even if a liquid market for patented ideas existed, valuation would be a very difficult problem. It is true that large companies often sign bulk cross-licensing agreements (for example, Sony and Samsung have cross-licensed a huge number of patents), but I have a hard time imagining a future where I can list a clever idea on EBay and sit back to consider bids from around the world.

The new predominance of intellectual property in technology industries is fed by a number of broader industry trends. First, IT and telecoms have become so complex that there is a greater willingness to accept the innovations of others. Gone are the days when vertically integrated firms handled every step of a product, from initial design to final sale. Now, a small army of specialist firms focus on narrow portions of technology, using intellectual-property rights to protect their inventions when they are licensed out.

Second, as many new technologies quickly turn into commodities, firms increasingly rely on innovation to remain competitive. Yet the return on investment in R&D is short-lived because more people innovate at a far faster pace than before. That means margins have shrivelled, explains Ragu Gurumurthy of Adventis, an IT and telecoms consultancy. “How to recoup the cost of innovation? By licensing the technology,” he says.

Third, customers are demanding “interoperability” and common standards rather than proprietary systems, which means different firms' technologies must work together smoothly. This often requires pooling patents or cross-licensing agreements.

Fourth, generating intellectual property is less capital-intensive than other aspects of the IT businesses because it relies mainly on people rather than bricks, mortar and machinery. That makes it attractive to many start-up firms. Venture capitalists often demand that firms patent technology, both to block rivals and to have assets to sell in case the firm flounders. This was particularly apparent during the internet boom in 2000. “In addition to the dotcom bubble, we had a patent bubble,” says Mark Webbink of Red Hat, a firm that sells Linux, an open-source operating system.

Companies cannot simply turn their back on what is happening in intellectual property. Even if they refuse to play the game, they may be unwittingly infringing someone else's patents because there are so many more of them around. Unless firms have patents of their own to assert so they can reach a cross-licensing agreement (often with money changing hands too), they will be in trouble. Thus many companies are acquiring large numbers of patents for purely defensive reasons, for use only to keep others' patent threats at bay.

...But when talking to executives in the technology firms themselves, the language you hear most often is that of “the arms race” and “mutually assured destruction”. Companies amass patents as much to defend themselves against attacks by their competitors as to protect their inventions. Many technology companies have recently championed reform of the patent system to deal with spuriously awarded patents, licensing extortion and massive lawsuits. “There is a broad recognition in the US that the patent system, if not reformed, will...begin to impede American competitiveness around the world,” says Bruce Sewell, general counsel of Intel, the world's biggest chipmaker.

This survey will argue that, despite such adjustment problems, the huge changes in intellectual property currently taking place in the IT sector will in time produce more efficient markets. But what do the IT firms themselves make of it all?

See also this article in the survey on IP in China and India.

The rise of China and India has mainly been underwritten by foreign companies, not indigenous ones, though this is starting to change. Both countries have been good at persuading firms setting up operations there to invest in training locals. Today, nearly all the large IT firms have big research centres in both countries, and local companies understand the need to develop their own intellectual property. Local people who went to Silicon Valley to find fortune are now starting up their own businesses in their home countries. Foreign venture capital is pouring in.

Without home-grown technology, India and China have to depend on foreign firms, and they do not like it. China, in particular, has seen a surge in the royalties it is paying to foreign firms, and is trying to stem the flow. When Qualcomm's boss went to China in 2001 to negotiate royalty payments for his company's third-generation mobile-phone standard, he agreed to accept less than what he charges others. Within a year, China was working on developing its own 3G wireless standard. If it succeeds, Qualcomm will see its royalties shrink further.

China and India have more to offer than just low costs, although these are clearly important. They are also able to deploy huge numbers of people to work on a project. Being able to throw bodies at a problem is vital in IT. It allows firms to do things such as speed up development cycles or explore alternative approaches that would not be possible with a smaller labour force.

In short, China and India are not simply taking over western IT jobs, they are changing the very process of IT development. It is not about doing the same thing cheaper, but about doing things that simply could not be done before. In that endeavour, intellectual property is becoming increasingly important.

There are limits to the optimism about India and China. Both countries have a culture of keeping technology to themselves. The western concept of patents is fairly new to them, and has proved controversial for countries at their stage of development. Also, both nations have huge institutional and infrastructure obstacles to overcome. Capital markets are embryonic. Big companies are coddled by the state. India's government bureaucracy is stifling; China's is opaque and corrupt. The legal system is uneven in India and consistently inadequate in China. Both countries badly need more experienced managers.

American technology executives with some experience of India and China are worried that the two are about to eat the rich world's lunch, but locals with deep knowledge of both countries think it will take at least a decade. Still, the overall trend is clear: the rise of China and India as centres of innovation will radically shake up the technology industry that is today based mainly in rich countries.

...Take Huawei Technologies, a big vendor of communications equipment, with revenues of $5.6 billion in 2004. This year, revenue from abroad is expected to surpass that from domestic customers for the first time. Around half of its 34,000 employees do R&D work, claims the company. Its patent filings almost doubled each year during the 1990s, though they have recently started to slow somewhat: the number this year will be around 2,400, and from next year it is expected to settle at around 3,000 a year. In 1995 the company created a special department to work on patents, which currently has 100 people on the payroll but will expand to twice that number next year.

“If you didn't have patents, you would be in a very disadvantaged position relative to your competitors,” explains Liuping Song, the head of Huawei's intellectual-property department, at the firm's headquarters in Shenzen. “Other companies approach you and charge you for using their patents.” So is the firm chasing after patents simply because other companies are doing the same thing? Mr Song laughs and says, “That is a difficult question to answer.” Then he adds: “We have to play by the rules of the game.”

Blog Archive