Among statisticians, economists, and business executives, “Big Data” is all the rage. Large and detailed data sets that, until recently, couldn’t even be stored on a computer are now managed and analyzed using innovative statistical techniques. Hopes are high that these advances will improve scientists’ ability to predict human behavior. Some enthusiasts even speculate that Big Data will render markets obsolete, enabling central planning of the economy. Big Data is more than a buzzword, but its potential is often wildly overstated. Continue reading “Why Big data Won’t Save Central Planners from the Knowledge Problem”
I’m going to start this post with a story that was quite common at my undergraduate institution, Western Carolina University, when I went there. You’re driving to campus for a class that starts in five minutes. You should have left earlier, but as a 20-something in college, you aren’t known for planning ahead.
Feeling hopeful, you pull into the parking lot looking for a spot. As you begin your search, you realize you’re in trouble. There are no spots to be found in the first lot! You frantically speed from parking lot to parking lot until you’ve covered virtually every square foot of your campus. Throwing up the white flag, you end up parking in that one lot in the middle of nowhere that’s at least a mile walk from your class, which results in you showing up awkwardly, rather than fashionably late for.
How can we solve such an issue? The students at WCU had no shortage of complaints, but very few solutions. Part of the answer can be found in the work of one of the two men awarded the 1975 Nobel Prize in Economics, Tjalling Koopmans.
The award was given “for their contributions to the theory of the optimal allocation of resources.” This may sound similar to some of our previous posts, and here’s why: These economists were awarded in the midst of what is known as “The Socialist Calculation Debate.” I won’t go too in-depth with its description, since it may be necessary for our posts with other laureates. Looking at you, Hayek!
However, the point is that the debate was over whether we could actually calculate the best way of divvying up scarce resources (often called “chalkboard economics). That said, anybody who contributed to that discussion would likely win a Nobel.
Koopmans focused primarily on the optimal allocation of resources when it comes to city planning, specifically traffic. He applied the basic ideas of optimal allocation to the transportation market.
“If cost is minimized in each branch of production on the basis of such a system of prices, each unit of any (divisible) factor of production will be used in such a manner that its contribution to the satisfaction of ultimate consumers is highest.”
Okay, that was a mouthful. What he means is that if we are acting competitively, i.e. behaving like a free market, every part of the industry will be geared towards making people as happy as possible.
In the same paper, Koopmans conjures huge brilliant theories at attempt to precisely coordinate the transportation sector by using marginal costs and benefits of the industry. Although they are definitely well thought out, they still fall short. So, without these massive theoretical undertakings to guide our transportation industries, how does the shipping industry, for example, effectively coordinate its transportation? Koopmans answers this, saying that perfectly competitive markets are closely tied to marginal benefit and marginal cost.
Who would have thought? The best way to imitate free markets was to allow free markets.
This idea is more fleshed out in a later book called “The High Costs of Free Parking” by Donald Shoup (Many thanks to Nick Zaiac for the recommendation). This came to the same conclusion with parking that Tjalling Koopmans did with shipping. When parking is heavily subsidized and made free to most, this messes up marginal costs and benefits, making it hard to analyze and correct inefficiencies. As a result, we have over-parking issues.
When examining the parking situation at WCU (or most college campuses), what we had was an over-subsidizing of our parking lots. Granted, I know the lots already had costs, such as the distance from classes, traffic jams, rambunctious drivers, parking fines, etc., but they were clearly not high enough with almost every student driving a car to school every day. We needed to increase the cost of parking so students would actually think if bringing a car was worth it, rather than attempting to create a free market situation through clever models.
Koopmans also had an interesting paper that discussed an issue that we’re still facing many years later, called “Measurement Without Theory”. He was speaking out against people gathering and analyzing data without having some general idea of what they’re looking for or why they’re looking for it.
With the advent of big data, many researchers and data scientists are succumbing to the idea that the data will just speak to them through patterns that just emerge from the data. Though patterns and trends may come about from large swaths of data, is it really meaningful without an underlying theory of what you think is happening? We can use data to observe, but meaningful research questions are exactly that: questions backed by some theory of how we think the world is supposed to work and some method of how to come close to finding the answer. For example, spurious correlations, though are pretty out there, illustrate the point of what you get when you seek answers without a theory.
I think this is a good warning to end on with Koopmans’ theories. He examined the effects and costs of transportation, and created brilliant models to try to recreate free markets. However, it’s important to note that doing this type of data collection without having some sort of mental framework is dangerous, and something we see in big data all the time.
If anything, at least we have a better idea as to why parking is such a hassle on college campuses. Focusing the debate on how to find the balance between parking spots and costs should prove helpful in reducing our parking frustrations when rushing to class.