oldleandude2

Insignificant Digits

Significant digits, the number of digits to the right of a decimal point that are warranted by the accuracy of the means of measurement, are a critical part of scientific investigation.   In developing products and services, the concept is essential.  For example, how many products have failed to “delight” customers because of tolerance stack-ups on component parts?  And how many of those components have failed to meet spec because the significant digits of the design did not match the capability of the production process?  In these cases, the number of places to the right of the decimal point is very significant.

But in many cases, the accuracy of our measures is insignificant.   Take, for example, the price of gas at the pump.  Why not $3.80 per gallon instead of the mixed decimal and fraction rendering shown in the sign to the right?   The answer is that in the 1930’s when gas price was 10-9/10 cents per gallon, (equal to about $1.40 per gallon today) a penny was worth ten percent of the price of the gallon – pretty significant.  The fractional price derived from an attempt by retailers to minimize the cost at a time of gross oversupply.  Today the condition is reversed; the third digit, and arguably even the second is not significant at all, at least not in a practical sense.

So it is with allocation accounting, another measurement system mostly developed during the Great Depression.   According to management accounting expert, Brian Maskell, “the bulk of current accounting methods formalized in 1930’s has changed very little since.”  A highly subjective cost accounting allocation scheme, for example, spread “insignificant” amounts of 1930’s overhead over labor cost.  Even then it was at best a rough approximation intended to recover the cost of non-production resources.. Today that once-insignificant figure amounts to a gigantic lever on relatively small labor amounts.

Overhead multipliers, once just small fractions of labor cost, are often two to three times the cost of labor today.  Call this an unintended consequence of the second industrial revolution.   In the words of Eli Goldratt, “traditional cost accounting is “precisely wrong.”   However significant the number of decimal points on a labor time may be from a scientific context, they are worse than insignificant in the bigger picture.   Yet many organizations continue to scrutinize value-added functions to four decimal places while ignoring huge non-value added wastes.   Make versus buy and pricing decisions based only on this model are as relevant as the 9/10¢ at the gas pump.   The difference is that gas pump prices, while confusing are essentially harmless, while traditional management accounting decisions are killing American business.

What other insignificant digit examples can you cite?  Won't you please share them.

O.L.D.

BTW  Brian Maskell will be in Boston on Monday & Tuesday, July 18 & 19 (that's next week!) to teach his sought-after "Lean Accounting for the Lean Enterprise" workshop - a must for CFOs and finance professionals from companies implementing an enterprise-wide Lean initiative. A few seats remain, so call GBMP at 617-287-7630 today to learn more and register or visit the GBMP website and click on Events: www.gbmp.org

This entry was posted in old lean dude, TPS, lean manufacturing, GBMP, Toast Kaizen, cost accounting, muda, kaizen, hoshin kanri, TPM, 5S, true north, lean accounting, poka-yoke, automation, optimization, toyota production system, inventory, Brian Maskell, made in america, Muri, shigeo shingo, value stream mapping, mura on July 12 , 2011.