The Delphi Technique
© Martin A. Armstrong
There are many forms of analysis and even more forms of mathematical forecasting equations — all determined to predict the future course of mankind. As discussed in the previous edition, judgmental forecasting still remains the dominant form of analysis employed not only by the financial community, but also throughout the corporate planning rooms around the world. Several well known case studies have determined that forecasting based upon human judgement is inherently riddled with error and holds the lowest rank in accuracy when stacked up against even the most inefficient quantitative model. Nevertheless, there is one form of judgmental analysis which has not been altogether that bad insofar as intuitive forecasting of technological advancement. This method is known as the Delphi Technique.
Although man has been trying to forecast his monetary future since the dawn of time, forecasting changes in our environment and technological advancement are relatively recent additions to this art of soothsaying. Personally, I can relate very well to the Delphi Technique after experiencing some of the problems it identified among professional expert opinions.
I have often been highly critical of the formal education process because it seeks to perpetuate the established views and discourages free thinking. The current upheaval taking place within the US educational system caused by the “Choice” approach should be welcomed with open arms. Yet at the educational level it is meeting with stiff resistance to change – not unlike the very teaching methods they employ. New students with degrees from the finest schools, MIT included, arrive at our doors like children believing in Santa Claus. Their training is merely an indication of discipline – not knowledge, and far too often they are merely vessels containing a collection of personal biases instilled within them by their professors. Economics, in particular, is tainted by those who confuse the science of economics with social reform – a particular characteristic of MIT. This is one problem that was also identified by the Delphi Technique.
The Delphi Technique was pioneered by the RAND Corporation. It is based on three characteristics: anonymity, statistical analysis and feedback of reasoning. The Rand Corporation found that in order to provide a forecast which would be semi–reliable, it was necessary to put together a team of professionals. They also found that anonymity was the key to future success. If the team were assembled and met together, naturally there would be some individuals with more experience than others. Clearly, those individuals would also perhaps be widely respected for their work. This caused problems within the team. Fresh, less experienced team members, would rarely challenge the opinions of more well–known experts. As a result, the opinion of the team would often reflect the opinion of the dominant team member and not a true unbiased consensus.
To solve this problem the Delphi Technique involved a team of experts who were kept in the dark as to who the other team members were. The team never met and acted without the influence or interference of other team members. Each individual ‘s forecast was taken and correlated. A consensus was then determined. Any individual who’s forecast fell outside the consensus was then asked to either defend his forecast with statistical evidence or rethink his conclusions. This process would be repeated several times until a reasonable hypothesis was concluded .
The Delphi Technique illustrated one key problem that plagues the financial forecasting industry today. It was found that the best ideas did not always surface from the most experienced member of the team. And as for those individuals who were perceived to be the best according to reputation — they were forced to support or conform. In an open group they would rarely be challenged due to their reputation and stature.
I must admit, that during the early 1970’s, 1 would often talk to many people about the possible direction of the markets. I would have my own opinion, but by the end of the day my confidence in my own work would decline. Only after finding that my own opinion would be correct afterwards – at least as often or sometimes more so than others – I realized that too much intermingling was not good. I decided that right or wrong I preferred to stand based upon my own work rather than someone else’s. If I was to be wrong, at least let it be for my mistake and not my peers.
A few years later, a very well known economist and central banker requested a meeting. I was in awe of his reputation and was surprised that he even knew who I was. He came to my office and spent the day. He pulled out his charts of economic movement since World War ll upon which he had plotted the course of short-term versus long-term interest rates. He then explained that every time short exceeded long, a recession soon followed. By now I was more relaxed and had slipped into an intellectual mode where I forgot about my junior status. I pulled out my charts on economic movement dating back to the 1700’s on short as well as long-term rates. I pointed out instances where short-term rates rose but failed to exceed long-term and still a recession followed. And as for the periods when short had exceeded long, there was no definitive correlation insofar as time and price were concerned. Short had remained above long for several years, while in some cases it was but for only a few months. There was also a lack of uniformity within the differentials. At times, short-term rates had risen to 3 times that of long-term rates, while at other points in time the differential would be less than 5%.
The conversation continued and by the end of the day I had converted him to my theories quite unintentionally. The following day the shock of what had transpired finally hit me – I had held my own against someone who’s reputation I could never hope to match. Yet had I been on a panel with this gentlemen, I would never have mustered the guts to launch an open challenge. Ironically, a decade and a half later I find myself in the reverse position at times when others are aware of my reputation and shy away from challenge.
The Delphi Technique is quite clever. Regardless of how high up the ladder you may think you are, you are still forced to back up your position rather than rest on your laurels. We are all only human, and with that comes the emotional baggage of clashing egos and insecurities. Some will argue and attack out of professional jealousy. Others will shy away from confrontation and fear looking like a fool. But once in a while, as the Delphi Technique discovered, the best analysis does not always emerge from those with the biggest reputation. Knowledge is not divinely endowed — it is earned. We will never have all the answers no matter who we might be. Someday, someone will always build upon your work. If free thinking is discouraged by institutionalized education, then progress is ultimately denied. Keynes himself was a free thinker. He would not have agreed with the current excesses of government intervention or fiscal irresponsibility. Yet his followers cling to his work in such an uncompromising method, they have tarnished the very essence of the man.