Competition as a discovery procedure

Econlog has an interesting little blog entry. It talks about how the US wireless scene, while appearing to be chaotic and disorganized, produces an unexpected result: we're on the verge of producing the best wireless solution available anywhere in the world, despite the disdain from Europe back when we were far behind their wireless solutions.

Basically, what happened is that our system, while appearing chaotic and disorganized produced a wealth of different solutions. None of which worked together. This drew the disdain from our European friends. But it's turning out that the wealth of products resulted in competition and now the offerings that we have are surpassing the rest of the world.

The conclusion? Competition is better than central planning at producing the best products and services. Of course, this result is really only unexpected by those who tout central planning.

I recently had another argument with a different co-worker. He was also in favor of central planning. In this particular case, the coworker was arguing that it was the job of our technology department to identify a single solution for a particular problem. I asked the question of why we had to standardize on that one solution to the exclusion of others? The answer was that we were trying to reduce costs in order to meet cost cutting requirements that our CEO put forth. And delivering a standard set of solutions (for everything) is the way that we reduce that cost.

Of course, if you don't measure all of the costs then you can produce a result that appears to cut costs. Putting in a single solution appears to take less time and effort than putting in multiple solutions. But when you have multiple solutions, the business can pick which ever one (or more) works best for them. Also, if you're built to support multiple solutions, it's not very hard to allow the business to make a recommendation on a new solution that you hadn't considered. Of course, this can get expensive. But what's not being measured when you throw out all those other solutions in favor of the single solution is the opportunity cost. What opportunities are we missing because the single solution we've chosen doesn't allow the business to function in the way they need to?

In other words, if we standardized on X, does that prevent us from selling to Y because X doesn't support Y? How much money did losing that opportunity cost us? I don't know. But neither does my colleague. The only way picking a single solution works is if we can accurately predict with a high degree of reliability all of the requirements that this solution will need to meet during it's lifetime. When has that happened ever? My colleague is betting that his single solution will meet most (if not all) of the future requirements. I'm betting that there's tons of stuff out there that he and I can't think of, and by standardizing on X, we lose those opportunities.

I would even go further and argue that this is an application of one of the most basic principles of economics: the law of supply and demand. By limiting the supply of available solutions, you increase the demand for those solutions. It may not be the best solution, but it's all that the business has, so it's what the business will use. The fact that there are no competitive forces to enable a line of business to go get some other solution, means that they're stuck with the solution that's there, even if it's not the best solution for them. The price of such a solution will never go down. Decreased supply => increased demand => higher prices.

On the other hand, if you create a ton of solutions, the line of business can make a decision between cost of the solution and the benefits it provides. Whatever happens, the cost goes down. If they pick the expensive solution because it offers a better benefit, it's because they decided not to pay the opportunity cost. This results in lower overall cost. If they pick the lower priced solution, they haven't incurred the opportunity cost - they're still doing the thing they wanted to do - AND they're paying less. This results in lower overall cost. Increased supply => decreased demand => competition => lower prices.

If you want to lower the cost, you have to increase the supply of solutions and let competition choose the winner.

I wish I could do a better job of communicating this to my colleagues.