Dear Mr. Kettle,
Years ago, when I was working in the manufacturing side of the computer industry, one of my more unenviable task was to improve something called the first-pass fallout rate (FPFR) of the products who’s manufacturing life cycles I was tasked with overseeing. For a bit of clarity to the uninformed, FPFR is a simple quotient of product that fails to meet basic manufacturing quality guidelines divided by the total production numbers for that product. It’s been said that for a manufacturing venture to master its profitability it must control its first-pass fallout rates. Whether that bit of wisdom still holds true in the decade since China joined the World Trade Organization, taking with it 5 million U.S. manufacturing jobs in exchange for guaranteed profit margins, is beyond my understanding.
I mentioned that the task of improving FPFR was unenviable for two very fundamental reasons. The first unenviable quality to this particular responsibility has to do with the concepts of statistical control limits (SCL) and acceptable quality limits (AQL), which are two concepts associated with understanding just how much of what one manufactures will actually reach the customers’ hands. Without getting into a dissertation in applied statistics, I’ll use a simple analogy that everyone can understand. If we flip a coin there is a 50% probability that it lands heads up. This is a concept collectively known and intuitively agreed upon. By extension, if we flip the same coin 100 times, we are guaranteed that roughly half of those flips will result in heads, for that is the statistical nature of that system. Of course, any 100 sequential flips of the coin may yield slightly more or less heads than the predictable fifty, which we refer to as standard deviation, as a metric for anticipating the unpredictable aspects that we collectively call chance. Still, as long as we don’t see sixty, seventy, or eighty heads out of a hundred flips, then we can be assured that we’re still rooted in the reality of three dimensional physics. Most of us will agree with the assertion, but a few folks will still be stubbornly compelled to test the experiment by flipping that coin all 100 times – perhaps even a thousand times, or more – for statistics and probability can be difficult concepts to grasp, much less believe. Manufactured goods follow a similar pattern of predictability. Through observation and carefully recorded metrics we know that any electronic component that can be properly evaluated within its operational parameters when first turned on and survive those tests will have a long and healthy life on the order of twenty to thirty years. We also know with reasonable certainty what percentage of those electronic components will fail during this initial evaluation from critical, major, and minor defects. As a result, we can take our understanding of these predictable patterns of failure and translate them into more complex systems, such that every resistor, integrated circuit, soldered joint, and printed circuit board trace can be tabulated into a predictable FPFR for any manufactured circuit board. By vetting any newly sourced part and managing the testing process, I maintained control of critical and major patterns of failure – failures that might otherwise have shut down my manufacturing lines or, worse yet, allowed defective product out into the world. On the other hand, minor failures associated with workmanship issues, required diligent monitoring of FPFR to identify and correct negative trends in human behavior – new employees, protocols changes, equipment maintenance modifications, etc. However, in an almost completely automated manufacturing process even workmanship issues were almost inconsequential. Still, there was always a certain percentage of predictable failure, based largely on the complexity of the product being manufactured. In other words, there was never an expectation of FPFR approaching zero, among those who understood the way in which this process worked.
On the other side of the FPFR coin was the need to reduce these numbers as paramount to profitability, for as I’ve already stated a manufacturing venture that masters its first-pass fallout rates masters its profit. Failure to reduce these numbers from one fiscal quarter to the next was looked upon unfavorably by management. As a consequence of this business axiom, the FPFR reports being submitted by the factories were already below statistical control limits calculated for the products, making the reports meaningless as tool for assessing the relative health of a product in production. Point in case, an overseas factory was producing 50,000 units a month of my most complex data storage system and consistently reporting an FPFR of precisely 2.0%, from one month to the next. Unfortunately, because of the complexity of the system, SCL predicted that the product could not be produced with an FPFR much lower than 6%, without a radical change in its design or the manufacturing process. In other words, the factory was suddenly coming up heads eight times out ten. Place your bets folks, if you’re feelin’ lucky! More accurately, they had long since abandoned accurate FPFR reporting, such that I could not reliably track workmanship issues, much less any other patterns of failures. This arrangement made for an interesting pattern of behavior toward the end of each fiscal year as the factory pushed its bone pile of defective product back through the factory line only to discover that the product still could not be qualified as ready for sale, spiking FPFR and causing company-wide panic that got me more than one uncomfortable audience with the division’s vice president. I did make several offers to pack my golf clubs and travel to this factory in order to observe what was happening and, perhaps, better understand the subtle discrepancies between what was written and what was real. However, there was talk about the office I might be disinclined to return to the States, if allowed to leave. I came to look at this time each year as my own personal Halloween, except instead of answering the door to find a gaggle of diminutive monsters yelling “trick or treat”, I found a bag of excrement ablaze atop my stoop, each and every time. Rather than scrapping and salvaging the defective product the company invested thousands of extra man-hours each year (engineers, quality assurance managers, electronics technicians, factory line workers, etc.) into what I came to refer to as the Frankie Resurrection Effort, simply because no one would accept the math that says man can only resurrect the dead within the confines of his own imagination.
I offer up this belated little Halloween tale not because it’s an anecdotal message about how we have a tendency to ignore the subtle complexities of mathematics out of ignorance or blindness. Rather, this tale was symptomatic of all the products that I worked on, in every factory around the globe (to be precise there were twelve products, in four factories, on three continents). The only consistency in the whole affair was the lack of consistency in the way in which FPFR numbers were fabricated from one factory or product line to another because few, if anyone, writing the reports understood the application of SCL and AQL as it applied to the products they were building. This eventually led me to pull the encoded manufacturing dates off the defective products, revealing that the defective products that suddenly spiked FPFR at the end of each business year spanned an entire year’s production. In other words, these were the dead bodies that didn’t make it off the line, were never counted and everyone refused to bury. They were the missing 4% that the math had predicted would fall from the production line. Before that gig, I had hair and it was not white.
While this may be intriguing stuff for an aspiring engineer, chomping at the bit for the day when manufacturing returns to our shores, I'm sure I’ve lost most of my readers. After all, this is a fable of mathematics that, while succinct, strives more for truth than fiction and, in so doing, loses its entertainment quality. Still, the moral of the story is simple enough to understand, for when we neglect the math the math will come back to bite us. Regrettably, we neglect the math all too often and, in my humble opinion, cost us millions of dollars a year in profit, in the aforementioned situation.
Another point in case might include the example you offered me recently regarding the City of Galveston’s employee retirement program. It’s regrettable that you had to reach back thirty years and half-way across the country to find a workable model for your example. After all, in the 1980’s, we had gross domestic product growth rate that was more than three times what it is today, for which many will argue as to the cause of its decline and who is to blame. We also had 5 million more good paying blue collar manufacturing jobs and Federal deficit spending unparalleled in the history of this country until the dawn of the 21st century. More to the point, as regards your assertions as to the primacy of annuities as preferable to Social Security, in 1981, AAA corporate bonds were paying annual interest rates four times higher than in 2012 and certificates of deposit were paying
36 times higher. What’s more, these rates outpaced inflation with a good 30-40% margin, such that once a retiree factored in the eventual cost of deferred income taxes there were still viable options for investing retirement funds, including annuities. In the last twenty years (that would be the last three presidencies) corporate bond rates and certificates of deposit have offered such low returns that once the spectre of deferred income tax is calculated into the equation, such investment vehicles barely pace inflation. That means anyone looking to place their retirement savings into an annuity had better have a sizeable chunk of money with which to start or be willing to make some concessions regarding the annuity schedule or how the annuity investment can be speculated. Something not really discuss in any detail regarding the events of 2008 is that had the investments banks been forced to resolve their short-term liquidity issues amongst themselves, it’s likely some would have sought the protection of bankruptcy courts in an effort to restructure their debts. I suspect annuity holders would have been among the first casualties to fall in this alternate strategy as they are, after all, little more than debt holders.
However, to get back on target, what’s more regrettable is that I have a bit of firsthand knowledge with non-profit organizations such as the City of Galveston. You see the state is littered with such organizations. On the other side of Houston, there was another non-profit business entity which also opted out of Social Security for its employees, at around the same time. However, moving forward to the late 1990’s, while this organization opted out of the Social Security system, it did not opt to channel any of its 6% payroll savings back to its employees in the form of higher wagers or increased contributions into its defined-contribution pension program. In fact, their retirement program was structured under the same ideology of high rates of returns that were part of the 1980’s, such that the contribution limits made by employees and employer met only the minimum legal requirements. As such, the frugal, diligent employee tracking his or her retirement savings in earnest, with even a modest understanding of the effects of inflation and compound interest, would also feel an obligation to seek the higher returns of equities investments (stocks) to ensure that even the modest rate of 3% inflation did not outpace their retirement contributions. However, the relatively anemic performance of equities in the first decade of the 21st century meant that diligent employee also had to dig deep into his pocket to offset unrealized gains or face insolvency in the event his retirement savings and investments proved inadequate. After all, there would be no Social Security benefits waiting for him at the end of the line.
Of course, this story has an ending, whether it’s good or bad is in the eye of the beholder. You see, with new Federal mandates in formulating distribution of Medicaid payments to hospitals, this non-profit hospital was about to lose a big chunk of its annual welfare check. To understand this predicament, you must first understand that Texas, like many other states, use to take the Federal Medicaid dollars issued to it and simply divide that portion earmarked for hospitals equally amongst them. If you were a hospital in a relatively affluent area, treating few, if any, Medicaid and uninsured patients, those funds were like a profit sharing check that padded your bottom line, even if you were a non-profit hospital. On the other hand, if you were a hospital treating a lot Medicaid and the uninsured, then you were just out of luck, ‘cause that’s how we do things here in Texas. But please rest assured that Texas also mismanages its own tax revenue with equal aplomb, having recently been directed by its own courts to restructure the
way it funds its school systems. More to the point, facing a significant reduction in Medicaid subsidies, beginning in 2011, this particular non-profit hospital was looking at immediate insolvency. As such, it was forced to sell itself or shutter it doors.
I would have found your example much more compelling had it simply been a little more current. Perhaps you could offer up something a bit more contemporary which reflects the math and economics of the time. For you see, whether it’s manufacturing or economics, a man must work with the system in which he finds himself, to the best of his abilities, even if he desires change for the better. After all, if a man finds himself standing in the streets of Rome, circa 100 A.D., it would be unwise of him to profess with loud defiance his love of Chinese food and Christian ethics, lest he sought to be drafted as the Emperor’s evening entertainment at the Colosseum.
Regrettably, I’ve been a bit more verbose than usual, but can find nothing to cut from what I’ve said, because it simply needed to be said. I do hope you’ll regal me with some more current investment advice, as I’m always keen to understand how others have successfully navigating a broken economic system in such dire need of repair.
Sincerely,
Mr. Pot
