Friday, July 12, 2013

Costs and Percentages

Last week, I talked about the percentages and perceptions. This week I want to be little more concrete in my example.

Let’s take a hypothetical situation, and say that 1 million US citizens suffer from something we’ll call Dihydrogen Oxide poisoning.   Let’s also say that an analysis of the situation says that for every 100 million dollars we spend, we can solve this problem for 90% of those still suffering from this affliction. That’s not a bad assessment, up front that sounds like that’s \$100 per person, or not a terribly bad investment.

Except for one little thing. That 100 million dollar investment only covers 90%.  So after the first 100 Million is spent, we’ve only solved the problem for 900,000 people, leaving 100,000 people still affected.

To get that remaining 100,000 people, we have to spend another 100 million dollars, but again, we’ll only solve it for 90% of them. So another 100 million spent, and we end up with a total of 990,000 people covered, and 10,000 people still left affected.

200 million spent, 990,000 people covered works out to an investment of a little over \$202 per person. But we still have 10,000 people left affected - it will take another 100 million to cover 90% of those remaining.

300 million, 999,000 people covered works out to an investment of a little over \$300 per person, but you still have 1,000 people left affected. But it’s not truly an investment of \$300 per person - the investment was actually \$111.12 per person for the first 900,000 people, and \$1,111.12 per person for the next 90,000 people and \$11,111.12 per person for the next 9,000 people.  It only averages out to a little over \$300 per person when you consider them all - and that’s reaching 99.9% of the people affected.

Are you willing to spend \$111,111.12 each for the next 900 people?  Or over a million dollars each for the 90 after that? Or over 11 million each for the 9 people after that? How much are you willing to spend to get that last person?

There is a point of diminishing returns, and you have to consider that when considering any policy. It might only cost a little bit over \$600 a person to cover 999,999 people affected by Dihydrogen Oxide poisoning, but it might actually be better to only spend \$300 per person to cover 999,000 of them, and spend that other \$300 million somewhere else, to provide relief for another problem.

Most public policy situations like this aren’t quite this linear, but they almost always have the same situation of a point of diminishing returns, which, unfortunately, is often difficult to tell where it is.   Also unfortunate is the mindset in government and public perception that we have to always try and cover 100% of the problem instead of recognizing the point of diminishing returns, and investing those resources elsewhere.  This is is unfortunate, because it is almost always the taxpayer who ultimately ends up funding bigger and bigger bureaucracies that ultimately cannot solve the problems at hand to 100% satisfaction. In fact, there often comes a point in which the size of the bureaucracy creates more problems than it solves. And yet, we continue to think that more is better, that we can solve it, when in fact, the better answer may be to recognize when we’re doing the best we can for now, and that further investment could be more effective elsewhere.

It may be distressing to think about it that way, particularly when children are involved.  But would you sacrifice 100 people to save 1 child? And then do that time after time after time, even after it’s been found that most of the time you still don’t end up saving the child? Most logical people would say “Of course not,” yet many of our public policies have reached the point they work just that way.

It makes no sense.

Friday, July 5, 2013

A Reasonable Percentage

Recently, I was a participant in a conversation online where one person was loudly criticizing what he perceived as a large number of problems with a company’s recently shipped new product line.  At first glance, he seemed to have a point - a fairly large number of people had reported significant problems with their orders.

But when the rest of the forum started analyzing the numbers of how many units were ordered vs the actual number of publicly reported problems, the problem didn’t seem too bad anymore - to most of us anyway. This large number of reports only translated to less than a single percentage point of the total number of orders shipped.

I’ve talked about numbers before, and the perception of them, but now I want to talk specifically about percentages, and our perception of them.

First and foremost, think about what you consider as an acceptable percentage of performance for yourself.  Are you a perfectionist that demands 100% performance from yourself and others?  or are you a little more relaxed, and recognize no one is perfect, thus finding yourself more comfortable with a lower number, like 99%, even 98% or 97% performance? Or are you more casual, and find no problem with a 90% or 85% performance rate?  Do you set higher standards for yourself than you do others?  If you expect 98% out of yourself, are you willing to accept less from others?

If you’re like most people, your expectations are probably pretty variable, depending on circumstance.  You might be an excellent cook who expects perfection in the kitchen, but you don’t have a green thumb, and are perfectly happy if you simply manage to keep the plant in your window mostly alive for another week.

It’s hard to put an actual number to things like that, but it’s a good mental exercise to do so, simply because trying to put a number to your own performance makes it easier to manage your expectations of others.  Because, quite frankly, in almost every circumstance out there involving other people, it’s difficult to expect 100% perfection, and even more difficult to actually get it.

A case in point - many years ago, I ran the warranty & repair department for a manufacturer of musical accessories. Part of my job was keeping the numbers that assessed the performance of the manufacturing line - they were expected to have no more than a 3% return rate on any piece of equipment they manufactured. Considering the manufacturing line was running anywhere from 500 to 1000 units a day, it meant the company expected to see no more than 15 to 30 units a day being returned for warranty issues.  For the most part, the production line was pretty good, usually keeping their return rate at just under 2%, a couple of times even dipping almost as low as 1%.  Even with those low error rates, the repair department still saw around 2000 units a year for warranty repairs, a number that always alarmed management until the percentages of production numbers helped put it in perspective.

Even though I was quite aware of the percentage of return from the production line, I was blissfully unaware of the return rate of my department and the standards it was going to be held to until one day my boss and I were hauled in front of the plant manager and ripped a new one for a warranty repair that had come back a second time for yet another problem.  We were the repair department, therefore something sent in for repair should never come back for repair again - that was completely unacceptable to the plant manager, and more importantly, to me.

That is, until my boss put it in perspective for both me and the plant manager.  You see, at that point, my department had processed over 500 repairs, and this was the first failure since I had taken charge.  That meant we had a 99.8% rate of success with our repairs. The plant manager still wasn’t happy about it, but it was pointed out that since there was no set performance standard for my department, it would thus be reasonable to assume that the standard should be the same as that for production - no more than a 3% return rate.

Ultimately, it was determined that the repair department should be held to a higher standard than production, and that standard was set at no more than a 1% return rate.  This first return was well within that rate, and the matter was dropped. Until, of course, another warranty repair came back a second time a few months later, and we were ripped a new one again, despite still being well below the previously established 1% return rate standard.  By the end of my three year stint there, the standard had finally become accepted, and when I moved on, the department had a less than 0.5% return rate for my entire employment there. Or, in other words, a 99.5% success rate.

We truly strived for perfection, but despite our best efforts, we couldn’t do it. But we did do pretty damn good, setting a new standard for the guy who took over from me.

The entire experience put success rates in perspective for me. While I may hold myself to a higher standard in the areas I am experienced in, I’ve come to realize that in most areas of life, anything from a 95% on up is pretty damn good.  Consider this - in most grading systems - an A grade is typically 90% or better. So even straight A students are given a 10% cushion for error.

The bottom line, and the whole point I’m trying to make here is that while striving for 100% perfection is worthy goal for an individual and even a company, it’s also an impossible task for a society.  We are not a society of straight A students.

If a law or policy is serving the needs of 95% or more of the population, it’s typically not reasonable to go to draconian lengths to try and reach that last 5% or less. It almost always puts an undue burden on the rest of society to achieve a goal that is almost never actually achievable. This is not to say we should never try to reach 100%, but I am pointing out that inevitably, we reach a point of diminishing returns, and as individuals and a society, we need to begin to recognize when we hit that point, no matter if it’s a law involving national security,  taxes or even children’s safety.

Friday, June 28, 2013

"We hold these truths to be self evident."

In just a few days, we will be celebrating the 237th birthday of the United States of America.

As almost every American knows, 237 years ago 56 delegates from the 13 colonies began to put their name to Thomas Jefferson's eloquently written Declaration of Independence, thus launching themselves and our country into the annals of history.

As you go about your personal celebrations and barbeques this Thursday, July 4th, 2013, take a few minutes of your time to re-read the Declaration of Independence, and reflect on its words.

And as you read it this year, also consider that the 56 men who put their names to that document were branded traitors by their government of the time.

If you need a copy of the Declaration,you can find it here:

http://uscode.house.gov/pdf/Organic%20Laws/decind.pdf

http://www.constitution.org/us_doi.pdf

http://www.archives.gov/exhibits/charters/declaration_transcript.html