With rapid development of programming tools and sophisticated trading platforms, a growing number of market players switch to the use of programmed trading. The article primarily blames quants for the use of programming and modelling tools in trading of derivatives. Whether it is correct or not, I am not sure. They surely developed a lot of new derivative instruments lately, many of which had been made possible thanks to very sophisticated models and computer-based programming. Some of the synthetic derivative instruments, say in option markets, can be analyzed and valued only with the help of the computer.
On the trading side, emotionless trading robots are pre-programmed to initiate trades when certain combinations of market conditions are realized. Panic or no panic, fear or no fear, euthoria or no euthoria - an algorithm-based trade takes place or is stopped. The presence of such non-human players in the marketplace is likely to amplify volatility and cause wild daily and weekly moves simlar to the ones we observe lately.
What's the lesson? It surely helps to learn what most common trading models are out there and make trading decisions accordingly. The second lesson is to avoid sophisticated derivatives. The recent crisis demonstrated that their valuation is fairly subjective and, ironically enough, is largely prone to the human error (on the programming and valuation side).