Wednesday, October 21, 2009

Blinded By Technology


Last night, the Frontline series on PBS featured a new program entitled The Warning. It was a revealing look at the long-percolating financial crisis and the role that one woman, Brooksley Born, played in trying to sound the alarm. This was not a story about the meltdown of 2007 - 2008. The program focused on the long-ago era of the Clinton Administration, which carried forward the previous Republican administration's posture on financial deregulation. In a nutshell (if there is such a thing when it comes to these complex events), a crisis was caused by the refusal of the Fed or the Treasury to consider regulating Over-the-Counter Derivative trading. Born, then the head of a minor federal agency called the Commodities Futures Trading Commission, tried to put forth regulations because she alone, it seems, saw that a meltdown in OTC Derivatives could cause a cataclysmic failure in the banking system. Her seniors in the administration, Alan Greenspan, Robert Rubin, and Larry Summers successfully crushed her efforts and neutered her agency. Within six weeks of the showdown in Washington between Born and Greenspan, Long Term Capital Management (LTCM) went into free fall on their derivatives business.

For those who don't know that much about economics, and I include myself in that group, LTCM was founded by the technology wunder kids of economics. They had developed highly sophisticated mathematical trading models that supposedly allowed them to make enormous bets with minimal risks. Some of their senior people were Nobel Prize winners in economics. For a while, it seemed to work and LTCM made terrific returns. Everyone wanted in on the action including over a dozen of the largest Wall Street investment banking houses. There was just one catch: the LTCM money machine was a Black Box and no one outside LTCM was allowed to see how they were making their money. It was a "Trust me or don't play" model. Worse yet, each of the banks was told that they had an exclusive deal with LTCM. When the walls came tumbling down because of a financial crisis in Russia in 1998, LTCM fortunes went south big time. Everyone wanted out and that is when the banks found out they had been snookered. In the end, the Fed and the Treasury called all the bankers together and told them they each had to pony up $300 million to buy-out LTCM and save the financial markets from imploding. The banks did what they were told and the crisis passed, but not without extreme anxiety both in Washington and on Wall Street.

The Frontline program focused on the need to regulate OTC derivatives - which remain unregulated to this day. There are now over $500 trillion in derivatives in the market. At the time of the LTCM fiasco, there were $17 trillion at risk. The stakes are obviously much higher today. Derivatives, of course, are only part of the problem that we are currently facing. Securitized mortgages and lax lending standards have caused an even bigger mess than we faced in 1998. But for me there is a technology story in all this and that is one of hubris. It is the folly to think that someone has figured out how to model the financial markets so perfectly that they can ignore unforeseen and unplanned events that can make the model so much worthless software code. No system as complex as world markets can be modeled today with that kind of accuracy. The arrogance of LTCM and the ignorance of the bankers who bought LTCM's Black Box is appalling. Yet, it shows the faith we are willing to place in "the best and the brightest" who seem to have all the answers...without any caveats.

Don't get me wrong. I am not saying we shouldn't try to build models to understand complexity, whether it is in financial markets or global climate models. My belief is that we should expose any model upon which very significant decisions are being made to outside review. As we now have the tools to tackle evermore complex problems, we need to remain aware that every model starts with assumptions. And you know the old saw about the word "assume"...it makes an Ass of You and Me.

If you missed the Frontline program, you can see it streamed on the web.

Wednesday, October 7, 2009

Another Great Newspaper Archive

While I am on the topic of archives, I came across another great historical newspaper archive. Penn State University has a wonderful collection of Pennsylvania newpapers dating from the Civil War era. Despite the seeming limitation in dates implied by the Civil War, I have found papers that date from the 1850's to the 1870's. The website takes a little getting used to but once you get the hang of it, it is a very powerful tool for searching and reading old papers. If you are curious as to what people were talking about back then, click the link above and check it out.

Monday, October 5, 2009

History at Your Fingertips


I have been away from this blog for a couple of months. I'm back. I have been spending my time doing some intensive reading and research into some of the history of 19th Century railroading. I have a few ideas that might mature into either a magazine article or a book.

One of the things that has been reinforced to me in my reading and research is how fortunate we are to now have extensive digital archives online. It is really extraordinary how much material that used to be available only in the largest university or public libraries is now readable online. I have been spending a lot of time in four archives in particular. The first is a combined effort of Cornell University and the University of Michigan called Making of America. The archive contains scanned copies of thousands of books and articles that were written in the 19th Century. The second is the Internet Archive which is a very rich source of text, images, and even motion picture clips. The third is a website that has ten million pages of 19th Century Upstate New York newspapers called Fulton's Postcards. The newspapers are all fully searchable and are a fascinating source of insight into the era.

The fourth, and in many ways the granddaddy of them all is Google Books. Not only are millions of books online and searchable but you can also save books to your own online library, make notations about the books, and get either original scanned images or digitized texts. If a book is particularly interesting and off copyright, you can download a complete pdf of the book to your desktop. To say that this is a powerful tool is an understatement. Google Books has some other very nice features including links to online booksellers and even the ability to find a library in your area that has the volume in question. If you are interested in looking into the history of any sort of technology, these are great to go beyond Wikipedia or just a straight Google search.

I have always been a used bookstore junkie and I have found some really terrific books browsing the shelves. But one of the nice things that comes out of Google Books is the link to used online booksellers that have let me buy books that I might never otherwise have found at very reasonable prices - sometimes as low as one cent (plus four bucks shipping).

Obviously, everything has not been digitized and the libraries and archives remain the first line of professional research. But for next to nothing, everyone now can access some of the best materials in the world.

There is probably nothing new in this post but sometimes I can't help but state the obvious. If you haven't seen any of these resources, check them out. If anyone knows of other archives I would like to hear about them.