Normally I would agree with someone who observes that criticizing without offering an alternative is just cynicism and not very helpful. Such an observation was implied to my recent post on the absurdity of the unemployment figures: “What alternate means of determining the actual employment figure would someone suggest?” In fact, there are cases in which that which the mistake being criticized should simply be stopped – the fact that there is no alternative to suggest does not mean you should continue making that mistake. Let me explain:
Suppose I am planning a surprise party for my brother who lives in Michigan. I am having the party at my house west of Chicago. It would be very, very helpful for me to know exactly what time he will arrive. I really, really want to know that number. The math is simple enough: Departure time + (miles driven / average miles per hour) = arrival time. It will be easy enough to know exactly when he leaves home and the number of miles is readily available with a great deal of precision. The problem is the average miles per hour. The fact that Chicago is squarely in the path makes that number unknowable. Depending on traffic, construction, accidents, weather and the presence of Chicago’s finest it can take anywhere from 30-40 minutes to 2-3 hours. For that matter, there are all sorts of things that can impact his rate of travel before he even gets to Chicago as he travels through Michigan and Indiana.
Now I can look at that reality and make one of two choices. I can accept the reality that there is no way of knowing the precise number and plan the event in a robust enough manner that it will succeed no matter when he arrives within a very wide range. Or I can simply decide that, because it would be very convenient to know the precise number and because I really would like to know an exact number, I can simply make one up and bet the success of the party on that number. Oh, it wouldn’t be purely made up. I would base it on my experience having made the drive a number of times, do a lot of research and create a complicated model based on reams of data, and I would rationalize the number based on the time of day and day of the year and how I think those things might affect traffic, weather and construction. It would still be a guess, of course, but I could fool myself into believing that, because I thought it out and have a rationalization for it that it is accurate.
By the way, it will also be a vehicle for me to blame the inevitable failure of the surprise party on something (or someone) else. How could I have known it would snow? That the tri-state expressway would be down to one lane? That an accident would take place?
The problem is that, as a boss told me in very clear terms several years ago, having an explanation for failure – even a very good explanation – is not the same thing as success.
Whether it is self-deception because decision making is easier if we have a precise number, or the allure of having a vehicle to put the blame elsewhere for bad decisions and subsequent failure, the latter course – pulling a rationalized number out of thin air and concocting a mathematical basis for it – is far too often the norm.
It would be very, very convenient to know the precise, total cost to make or do something. It would make decision making easy to know the exact financially optimum quantity of something to make at a given time. Neither can be known. They are all wrapped up in costs that are, at best, indirectly related and costs that vary with time, rather than volume. How much of those costs has to do with any specific product depends on capacity being used, where the constraints are in the process and what else the business is doing. Those numbers are only derived when fictitious numbers are plugged in where reality is unknown. The fact that the ERP systems and pricing models would be really, really cool if the numbers were accurate doesn’t make them so.
Same is true for the politicians and economists. Employment figures and national manufacturing productivity are unknown and unknowable. The fact that it would really, really, really be helpful to know the employment and productivity rates doesn’t make the wild-ass-guesses and rationalizations accurate. The 90% confidence level in the numbers should have the Six Sigma aficionados out there apoplectic.
So if the numbers are no good how are we to make decisions? You do so using the numbers as, at best, directional indicators or as the boundaries within reality lies, but the decisions themselves come down to intelligence, wisdom, experience and values. You try to make decisions in a robust enough manner that the course you chart can tolerate some variation. The problem, of course, is that it means the decision maker is responsible for his or her decision.
The alternative is to accept the inevitable failure resulting from making decisions based on well rationalized but known bad numbers, and relying on the numbers to provide that ‘very good explanation’ needed to shift the blame for the failure elsewhere.
So to the question, “What alternate means of determining the actual employment figure would someone suggest?”, or what better means of calculating standard costs or batch sizes can we suggest, I believe that is the wrong question to be asking. The correct question is how can decisions be better made without relying on inaccurate economic data, or inaccurate standard costs and batch size math?