I was sitting in a traffic jam one day, and a politician was being interviewed on the radio. He was arguing for spending a sizeable sum to improve the very highway on which I sat motionless. The interviewer pointed out the large expected cost of the project, and noted that it would actually make traffic much worse while it was under way.
“Ahh yes, but it will be well worth it”, the official asserted. “It would shorten the average commute by at least 10 minutes for thousands of commuters each day. That translates into more time spent on the job. That increase in productivity is worth billions of dollars a year!”
This is a classic example of a frequently employed logical fallacy – that small changes aggregated over huge numbers of incidences add up to big impact. Multiplying the few minutes saved times thousands of people times 200 days a year times $x per hour does indeed yield a big number, but is it really “saved”?
While it is logically possible to measure the average commute before and after the road work, and thus verify the savings for various commuters at various times of the day, how could the total aggregate savings possibly be measured or verified? What if, instead of working an extra 10 minutes, most commuters spent an extra 10 minutes in bed before leaving for work?
Over the years, I have seen countless examples of phantom benefits. A printer salesman once told the CIO of a big corporation that because the salesman’s printers were 10% faster, the company would save “millions on the bottom line” because employees would spend less time standing by the printer waiting for their output. He even went so far as to cite the savings in $ per share. But how could that ever be verified? The micro impact may be verifiable, but extending that to an aggregate overall benefit is a huge leap of logic and faith.
To be sure, there are many examples of how small effects can be aggregated over large incidences to yield tangible, measurable, verifiable results. Anyone who has seen what is left of Plymouth Rock knows full well the effect of thousands (perhaps millions) of tourists taking tiny pieces of the rock for souvenirs over the years. Many other famous antiquities have suffered similar fates, including Stonehenge, the Acropolis, the Roman Forum, and others. Incidentally, in the above printer example, the CIO approved the purchase because the new printers used less ink – an impact he could see directly in the annual ink budget.
Today, phantom economic claims are everywhere, particularly in the debate over the best ways to combat global climate change. For example, there can be no doubt that increasing the gas mileage of cars leads to lower emissions – measurable, verifiable, and undisputable. But does that translate directly into fewer particulates in the atmosphere? What if drivers drive more miles once they know that they can go farther for the same investment in fuel?
As a rule, if you cannot follow the micro effect through to an identifiable macro impact, then the macro savings are “phantom”. In no way does that mean that the micro level impact is worthless. A ten minute reduction in commuting time, or a 5% improvement in gas mileage, or a 10% reduction in printing time may be absolutely verifiable and of great value to individuals. But claims of gargantuan benefits at higher levels are rarely supportable. That said, if 1 million people read this blog and stop accepting dubious claims of huge benefits from small actions, then the value to society would be incalculable.