Here’s a long post from Megan McArdle on health care and it’s well worth reading.
In the Wall Street Journal, Randi Weingarten of the American Federation of Teachers, writes that Markets Aren’t the Education Solution.
Market-based reformers advocate using student test scores to evaluate and compensate teachers, increasing the number of charter schools, firing teachers in low-performing schools, and relying on corporate executives and business practices to run school districts.
But there is increasing evidence it doesn’t work.
Using test scores to evaluate and compensate teachers is not market-based reform. It’s test-score accountability.
As a market-based reform supporter, I agree that it makes little sense to use test scores to judge teacher performance. Test scores should be used to judge student performance.
Also, relying on corporate executives and business practices to run school districts are not necessarily market-based reforms. But, they may not be bad ideas.
The distinguishing characteristic of a market-based reform is that it gives more power to choose to the end users of the education system — the parents of school age children. Any reform that does not clearly do that should not be referred to as market-based reform.
One conversation from the dinner table this weekend sticks out.
The topic was a ranking of favorite candies in Readers Digest. The #1 favorite candy was Reese’s Peanut Butter cups. Some found this surprising. They didn’t care for peanut butter cups and they didn’t think a majority of people they knew were fans of peanut butter cups either, so they couldn’t figure out how Reese’s PBC were #1.
I must have been tired.
I didn’t have the motivation to explain that in a market with many choices (many choices usually result in a free market), a majority following is not necessary for a #1 ranking. With so many candies to choose from only a small percentage following, sometimes in the single digits, is required to earn the top spot.
I’ve encountered this before, where folks confuse #1 and majority. I call it the majority fallacy, even though it’s not exactly the same thing as the real majority fallacy (which says that just because the majority of people believe something, that doesn’t make it true).
In a business process I work with, a market survey often comes back to say that such-and-such is the #1 problem. Based on this survey, the business designed ways to solve this problem and each year they were surprised to see how little effect their efforts had.
They never seemed to understand that even though the problem was #1, it was only cited by a relatively small percent of customers. They thought they were going after a “majority” problem, when they were really attacking a minority problem.
Another common mistake made by bureaucratic organizations in innovation or testing new ideas is too much planning.
Large organizations tend to be made up of people with Type A personalities. This is fine and good for the parts of the organization that have already passed their market test and are producing profits. You want well-oiled operations in order to efficiently deliver existing value propositions and Type A folks are good to make sure the all the t’s are crossed and i’s are dotted to achieve that.
Unfortunately, Type A folks often get too heavily involved in the innovation process in large organizations. This leads to over planning, which suffocates innovation.
These folks will plan the execution of the new idea to a tee before any actual market testing happens. Their plans will look impressive to other Type A folks and these plans are often the output that the organization’s managers reward on.
When the idea eventually reaches the market and doesn’t perform as planned, they tend to pull the plug on it without giving it a chance to adapt and respond to customers because they’ve scaled it out so large that it’s too expensive to keep rolling. It’s a one-shot trial.
Large organizations can learn something from the real new idea market – entrepreneurship.
Don’t over plan. Test small. Keep Type A people away from the innovation process until the others hit upon value propositions that customers are willing to pay for. Get the innovators away from HQ. Let them adapt their ideas based on real world operations. Give it a little time and patience. If you keep it small, costs won’t be a big concern.
Keep in mind that nearly all successful businesses started small, adapted for awhile through trial-and-error experimentation until they hit on a winning formula that got a good customer response and then expanded almost on a self-funding basis.
Arnold Kling and Nick Schulz write about a common barrier to innovation in big companies in their book From Poverty to Prosperity (p. 188):
To control intrapreneuring [development of new ideas within a company], corporations set up bureaucratic filters through which new ideas must pass. The bureaucracy is designed to kill most new ideas, because most new ideas offer poor return on investment. Corporate decisions are made by committees. In a typical committee, no individual has the power to say “yes” to a new project. On the other hand, almost every member of a committee has the power to veto a new project.
Observers of organizational behavior have noted that in committees one is more likely to be regarded as intelligent and a good team player by one’s peers by arguing against a new idea than by arguing in favor of it. Middle managers who fight for new ideas are regarded as troublemakers, even if they succeed in convincing corporations to undertake the projects they propose.
A corporate middle manager who fits Pinchot’s description of an intrapreneur is likely to be driven to leave a large organization to start a new enterprise as an entrepreneur.
Here’s a different idea for these big companies: Give it a try.
Don’t vet the idea in a conference room. Vet it in the marketplace. Go small and cheap. Give the project to a small group of people and get them out of headquarters and let them adapt it to real world feedback.
I’ve been posting enough recently on failure, that I have decided to add it as a category for this blog.
The Economist article reinforces what I have written about, that failure is necessary to have a chance at success and that the stigma on failure is unproductive.
It also makes a distinction on failure that I’ve been thinking about lately as well.
But simply “embracing” failure would be as silly as ignoring it. Companies need to learn how to manage it. Amy Edmondson of Harvard Business School argues that the first thing they must do is distinguish between productive and unproductive failures. There is nothing to be gained from tolerating defects on the production line or mistakes in the operating theatre.
That’s a good distinction to make. The article gives this example of unproductive failure:
James McNerney, a former boss of 3M, a manufacturer, damaged the company’s innovation engine by trying to apply six-sigma principles (which are intended to reduce errors on production lines) to the entire company, including the research laboratories.
I’m not sure I’d agree with that. With 20/20 hindsight, this appears to be an obvious unproductive failure. But, I’m not sure the impact of six sigma on innovation was well known prior to McNerney’s trial and from that trial we all learned something — that six sigma is a great tool for consistency in operations but not so good where we need leeway for experimentation. Thanks Jim.
I think a good example of an unproductive failure is running up credit card debt to buy clothes, cars and electronics and then declaring bankruptcy. Or promoting a high school assistant football coach to a head coaching job in the NFL. There’s a very small chance that some form of success could come from these actions — but it’d be a fluke.
While I’m not sure there are any rules to distinguish upfront a productive and unproductive trial, I do believe the distinction is a good idea to keep in mind when trying something.
At the very least you should be asking if the trial, or something like it, has been tried before and what the results were. If it was a failure, then it’s good to think about why and why you think your trial may not be.
This won’t always be definitive and prior similar failures do not need to keep you from trying again, but considering such things can help set expectations and keep you disciplined to staying with a low risk trial. Many times I’ve seen people, convinced of their own brilliance, bypass the trial phase (or worse, ignore or misread the results from the trial) to invest heavily in something to become a hero, usually becoming the goat.
From the Wall Street Journal: Amazon to Add Library Lending to Kindle