Ali’s blog

Mostly quant stuff with occasional digressions

Archive for December, 2007

The changing structure of employment in the US — and its likely impact on quant work

Posted by alifinmath on December 30, 2007

Here is the link to an interesting piece in WorldNetDaily:

http://www.worldnetdaily.com/news/article.asp?ARTICLE_ID=59398

Below are some excerpts:

While the price of a college education has skyrocketed far faster than inflation, many careers for which colleges prepare their graduates are disappearing. U.S. News’ Best Careers guide concludes, “college grads might want to consider blue-collar careers” because bachelor’s degree holders “are having trouble finding jobs that require college-graduate skills.”

Where did the higher-skill jobs go? Both large and small companies are “quietly increasing off-shoring efforts.”

Ten years ago, we were told we really didn’t need manufacturing because it can be done more cheaply elsewhere, that auto workers and others should move to information-age jobs. But now the information jobs are moving offshore, too, as well as marketing research and even many varieties of innovation.

The flight overseas includes professional as well as low-wage jobs, with engineering jobs offshored to India and China. Thousands of bright Asian engineers are willing to work for a fraction of U.S. wages, which is why Boeing just signed a 10-year, $1 billion-a-year deal with a government-run company in India.

But it doesn’t make sense for parents to mortgage their homes, or for students to saddle themselves with long-term debt, to pay overpriced college tuition to prepare for jobs that no longer exist. Tuition at public universities has risen an unprecedented 51 percent over the last five years.

A Duke University spokesman said that 40 percent of Duke’s engineering graduates cannot get engineering jobs.

Now let’s dwell on quant work for a moment. There is no good argument why quant work cannot migrate to lower-cost locations in the same manner as programming work. In fact, there are already quant shops in India that handle outsourced quant work. I expect this trend to accelerate. The choice between a $200,000 quant in the US and a $40,000 quant in India is a no-brainer. But let’s see what the experts say:

“For Citigroup to do high-end work in India is a no-brainer,” says Ravi Aron, senior fellow at Wharton’s Mack Center Mack Center for Technological Innovation. “Citigroup has been doing bond pricing and investment banking work in India for a while.”

In the initial stages, much of the outsourced work was at the “data level,” says Aron, adding that the

business model matures as companies get accustomed to working with their offshore partners and “see that there is a lot more they can do.” That opens the possibilities of outsourcing higher-end assignments such as bond pricing, equity research and financial market analytics.

“Whenever work involved number crunching, quantitative analysis, mathematical formulation, statistical data analysis and numerical calculations, managers in India, Singapore and China would say this is low complexity work,” says Aron. “They could find the people to do those jobs and deliver quality on scale, month after month, year after year.

What I’ve quoted above doesn’t say anything I don’t already know: this is the shape of the future. It’s not just that jobs are moving offshore, it’s that the financial industry is in the throes of major structural changes — and shall remain so for the indefinite future. For instance, the growth in overall employment of quants over the last so many years has probably come to an end. And companies that do need quant work done will probably look ever more keenly at offshore options. Were I a manager with bottom-line accountability, this is certainly what I would do….

Anyone thinking that a US degree will land them a lucrative career in the US is probably grossly mistaken. Even if a lucrative job is available, it may not stay on American shores: there’s no logical reason it should. And I’m talking now of graduates from the best quant programs: Carnegie-Mellon and Haas, for example. For others it will probably be worse.

Posted in Uncategorized | Leave a Comment »

A body blow to US science funding

Posted by alifinmath on December 29, 2007

http://fire.pppl.gov/doe_budget_2008_%20nature_122407.pdf

“This is probably the worst budget for science that anyone can remember,” says Michael Lubell, a spokesman at the American Physical Society in Washington DC. “It absolutely devastates and probably wipes out American high-energy physics.”

 

Basic research is critically dependent on state funding. Private-sector funding is generally for specific commercially-targeted work and tends to be development rather than fundamental research, which is a dicier proposition altogether — but with enormous social and economic paybacks if and when it succeeds.

Other major economies are ratcheting up their basic research funding while the US is cutting back on its own. The country is showing all the symptoms of an incipient banana republic: lack of investment in human capital, lack of investment in basic research, lack of investment in infrastructure, a widening chasm between rich and poor, persistent deficit financing, gargantuan trade imbalances, and excessive spending on the military being but some of the symptoms.

Posted in Uncategorized | Leave a Comment »

Donald MacKenzie on option theory

Posted by alifinmath on December 28, 2007

MacKenzie is a professor of sociology at Edinburgh. He has put forward some very interesting ideas about the role of option theory in creating the very option market it is supposed to be describing. He came out with a book last year — An Engine, Not a Camera — published by the MIT Press, where he discusses this at length. There’s an earlier paper of his — Option Theory and the Construction of Derivatives Markets — published in 2005, where he argues the same thesis; it’s available here. Some excerpts:

“Implied volatility” is an inherently theoretical notion: its values cannot be calculated without an option pricing model. By simplifing the options markets’ complexity to a common metric, “implied volatility” allowed the burgeoning trading firms of the 1970s, such as O’Connor and Associates, to expand and extend their activities by co-ordinating teams of floor traders operating on geographically dispersed options exchanges. What was being traded on these exchanges, the firms reasoned, wasthe Black-Scholes-Merton model’s fundamental parameter: volatility.

The idea that Black-Scholes-Merton might have created the very market structure it was purporting to describe:

In the case of the Black-Scholes-Merton model, the claim of Barnesian performativitywould thus be the claim that the market practices informed by the model altered economic processes towards conformity with the model – for example, altered patterns ofmarket prices towards what the model postulated – and the model was thus an instance of  “knowledge substantially confirmed by the practice it sustains.”

That this might be the case is suggested by the way in which the discrepancies between model and market seem to have diminished rapidly in the years after the model’s publication in 1973.

And why B-S-M was adopted when its initial correspondence with numbers and practices wasn’t initially that good:


Why might an options market participant in the 1970s have chosen to use Black’s sheets or another material implementation of the Black-Scholes-Merton model? The answer might simply be because the sheets were a good guide to market prices, but, as noted above, the fit between model and market was not always close, especially in the earlier part of the decade. Although it is difficult to be certain of the reasons for the dominance of the Black-Scholes-Merton model, a number of factors seem likely to have been significant. One factor – perhaps the factor closest to Bourdieu’s emphasis on the inter-relations of language, power, legitimacy, and cultural hierarchy – was the authority of economics. Financial economists quickly came to see the Black-Scholes-Merton model as superior to its predecessors. As noted above, it involved no non-observable parameters except for volatility, and it had a clear theoretical basis, one closely linked to the field’s dominant viewpoint: efficient market theory. The Black-Scholes-Merton model thus “inherited” the general cognitive authority of financial economics in a political culture in which economics was a useful source of legitimacy, and in which, in particular, the status of financial economics was rising fast.
Chicago floor traders in general were and are not in awe of professors. From their viewpoint, however, the model had the advantage of “cognitive” simplicity. The underlying mathematics might be complicated, but the model could be talked about and thought about relatively straightforwardly: its one free parameter – volatility – was easily grasped, discussed, and reasoned about. Kassouf’s model, in contrast, was a regression equation with six coefficients that required econometric estimation (Kassouf 1965, p. 55). An options pricing service based on Kassouf’s model would perform the requisite calculations, but from the user’s viewpoint such a model was a black box: it could not be reasoned about and talked about in as simple a way as the Black-Scholes-Merton model could. The many variants of, modifications of, and alternatives to Black- Scholes-Merton that quickly were offered by other financial economists also had a crucial drawback in this respect: they typically involved a mental grasp of, and estimation of, more than one free parameter – often three or more. Another factor underlying the success of the Black-Scholes-Merton model was simply that it was publicly available in a way many of its early competitors were not. As U.S. law stood in the 1960s and 1970s, it was unlikely that an options pricing model would be granted patent or copyright protection, so there was a temptation not to disclose the details of a model. Black, Scholes, and Merton, however, did publish the details, as did Sheen Kassouf (whose model was described in his PhD thesis). Keeping the detail private may have been perfectly sensible for those who hoped to make money from their models, but it was a barrier to the adoption of those models by others.

And how adoption of B-S-M created a closer correspondence between it and the market:

As noted above, with plausible estimates of volatility the Black-Scholes-Merton model tended to generate option values that were below the market prices prevalent in the ad hoc put and call market and in the early months of the Chicago Board Options Exchange. For a critic of the model such as Gastineau, that was an indication that the model undervalued options. However, it also meant that market competition tended to drive option prices down towards Black-Scholes values.

Black-Scholes prices were, in a sense, imposed even upon those writers of options who believed such prices to be too low: they either had to lower the prices at which they sold options, or see their business taken away from them by the adherents of Black-Scholes.

Other reasons for adopting B-S-M:

Assessing the risks being taken by a trader was far from simple: he or she might hold dozens of option positions, and perhaps positions in the underlying stock as well. The Black-Scholes-Merton model’s deltas could, however, be aggregated to a single measure of exposure to the price movements of a given stock. If a trader’s aggregate delta was close to zero, his or her positions were “delta-neutral” and could be considered to a first approximation well-hedged; if the delta was substantial, then his or her positions were, in aggregate, risky. Sophisticated risk managers learned not to stop at delta, but also to consider the other measures colloquially known as “the Greeks,” such as gamma….

Posted in Uncategorized | Leave a Comment »