Allan

ASA

Allan Sayle Associates

for business improvement services

ASA

Do statistics show quality improvement programs damage national gdp?

By Allan Sayle

Summary: Statistics make for a useful servant but a poor master. Blind application in process control or any other application can lead to strange conclusions

--------

It is often popular to remark derogatively of statistics their credibility follows after lies and damned lies. I would contend, though, that far from their residing being in the basement of reliability, recent years events prove there is a species even worse: corporate accounts. Of course, in a supposedly post-Enron world those in the USA, for example, are supposed to think Sarbanes-Oxley will elevate accounts relegating 'statistics' to their traditional position in the Hades of trustworthiness.

"With statistics you can prove anything", goes another oft-quoted pejorative remark. Perhaps.

Statistics must always be interpreted with care. In most training courses on the subject, one generally learns in the very first lecture that one of its most humble members, the average, can be highly misleading. A commonly used demonstration of that fact being one related to earnings, thus:

If in a two-citizen country, one citizen earns $1,000,000 per year while the other only earns $1.00 per year, the average citizen's income is $500,000.50 per annum.

Of course, our $1.00 friend will be unlikely to secure a bank loan to buy a new Maybach motorcar on the basis of his being a citizen of that country and his actual income!

One sometimes encounters a silly use of statistics. Take, for example, the recent aspiration of an Austrian gentleman who is trying to build his hedge fund. Apparently, his goal is to make sure all citizens can be richer than the average. At least, that is what one surmises reading his alleged statement, "Why shouldn't everyone be able to make above-average returns in any market?" (Ref. 1) Recalling literature’s famous Lake Wobegon, where all children did better than the average, perhaps his fund should be called "Wobegon Investments". My view is that anyone possessing such a grasp of mathematics is not to be entrusted with my savings for it would likely be a case of "Woe they're gone" maybe with him, too! Perhaps he has a framed photo of Charles Ponzi on his desk? Ahem!

Recognizing the maxim caveat reader when noting economic statistics, I am naturally cautious when informed plant utilization rates in America are risen to levels "close to those seen in 1999 2000...while production capacity continues to grow modestly." (Ref 2.)

My experience tells me that while factories may be busy and appear to be using more of their capacity, a significant proportion, affecting results, is probably producing waste. It is not how much capacity is being used that is as important as how much useful, value-adding output is being achieved. The ratio of output to input defines efficiency. And it is towards improving that simple ratio that a business' quality program is dedicated. Indeed, one may be better pleased to note one’s client has a lower capacity utilization after jointly working on a value-added program.

Economists, of course, may shudder at the thought. More so since, as that factory then consumes less raw material and energy, its suppliers' outputs are reduced and because the waste disposal trucks and recyclers, whose business rely on someone's "hidden factory" producing useful waste, are using less of their capacity. Put together, national industrial activity is lower as a result.

And so, with tongue in cheek, one might conclude the end result of quality improvement programs is a reduction of gross domestic product, gdp, which is another statistic to be viewed with care. Small wonder, then, top management and political leaders do not give those programs their fullest support.

If, then, we want to improve our national gdp, and create jobs thereby perhaps we have a duty to that important stakeholder - our national community - to encourage the production of as much waste as possible. But, when I look at a lot of consumer goods on store shelves maybe that is already being done.

It is unwise to base one's conclusions only on one type of statistic. We must always remember that when considering process control. Though numbers can set you free, they can mislead if one does not understand the bigger picture to which they are supposed to relate. Caveat reader?

References

Business Week, March 6, 2006, page 80.
Business Week March 13, 2006, page 28.

March 16, 2006

Allan J. Sayle

Associates and professional colleagues

Useful sites

Global resources