Putting Microeconomics to Work

I’ve long been puzzled by the almost complete disconnect between real-world businesses and academic economics.  After I graduated from college, I went to work as a management consultant.  Almost nothing I learned as an economics major proved helpful to me in that job.  Then, when I went back to get a Ph.D., I thought what I had learned in consulting would help me in economics.  I was wrong about that as well!

Ever since, I’ve felt that both business and economics would benefit from a greater connection.  Why don’t businesses set prices the way economics textbooks say they should?  Why are randomized experiments so rare in business?  Why do economists write down models of how businesses behave without spending time watching how decisions are actually made at businesses? The list goes on and on.

It’s taken a while, but the business/economics connections are finally starting to happen with greater regularity.  John List and I wrote an academic piece about field experiments in businesses a few years back that focused on how partnering with businesses could help academics with their research.

The benefits are also going the other way.  The Economist has a nice article about how microeconomists are adding value to businesses.  (I’m sure the economists mentioned in the article are delighted to be included; I’m almost as sure they will hate the cartoon likenesses that accompany it!)

For what it’s worth, I’m trying to do my part to improve philanthropy and business through a little firm called The Greatest Good.  But, damn, it turns out to be a lot harder to make things happen in the real world than it is in the ivory tower!


Tell me about it!

H. Bell

Two issues:

1) Economic models are normative by nature and don't necessarily reflect how things really are. Plus they're simplistic (all other things constant) and don't add the additional layers (behavioral, environmental, cultural) that go into decision-making. For example, one place I worked we always asked "What are the 'people' implications of this change" after we did the analysis.

2) Like informationally efficient financial markets, by the time the microeconomic model recognizes the need for an adjustment (like price), the adjustment has already happened. The models lag and are often built on historical data.


Kind of like "Global Warming" models . . . not so good at forecasting, but really good at hindcasting . . . after adjustments for historical data (reality).

Michael Cohen

This is actually rather straightforward, isn't it? Business owners stick with what they "know" works, because the cost of experimentation is perceived to be too high for the value that they think it will return. Even if they are wrong about both the costs and the value.


Randomized controlled trials are conducted on a pretty massive scale when it comes to e-commerce and web useability. E.g., should the 'Buy Now' button be green or red? The answer is often determined by serving different users different pages (randomly) and gathering statistics over the course of days or weeks. The winning button gets implemented. You have probably personally participated in such experiments without ever knowing it.

Enter your name...

Good advertising firms also do RCTs to find out which ads work best. That's how we learn things like the more text, the more sales, even though very few people read all the text. If we didn't test, you wouldn't be getting multi-page letters from charities, in black-and-white "typewriter" font with "handwritten" circles and arrows and notes scrawled in the margins in "blue ink pen" color.

John Van Reenen

Business & academia often don't connect http://www.freakonomics.com/2012/11/27/putting-microeconomics-to-work/ But here is a shot http://hbr.org/2012/11/does-management-really-work/ar/1


as T. H. Huxley put it:

"The great tragedy of Science — the slaying of a beautiful hypothesis by an ugly fact."

It has taken several hundred thousand years for science to take hold of human imagination precisely because nearly all of us prefer beautiful theories to ugly facts. The preference for beauty over truth is so prevalent I only wonder at the exceptions.


Academia and reality differ? This is news?


Levitt is elaborating on the relationship between theory and practice -- which can be stated thus:

-- In theory, there isn't any difference between theory and practice.
-- In practice, there is.

Paul S Vigil

It's always been about politics hasn't it?


Pricing in particular, involves much more consumer psychology and partner relationships than basic economic models. Unfortunately, many businesses are very failure adverse (Silicon Valley is the most noted exception) and employees are made to fear taking risks which causes people to view trials as a sign of weakness. Many business leaders would accept a good-enough outcome and be seen as a successful, decisive leader rather than admitting what they don't know and finding the best outcome.

frank reagan


In Spanish we say: "Everything is in the books".

And it is, indeed. Read Von Mises' "Human Action" and you'll start to understand why economic models doesn't work in the real work with real PEOPLE making real decisions.


Dr. Constantinos Charalambous

I agree with the normative nature argument. In fact, one can argue that there is no such thing as objective theory as economic theory is based on assumptions which are themselves subjective. To illustrate this I wrote a piece about "the theory of relativity in economics" . Feel welcomed to take a look here:


John F. Opie

Hmmm...I work as an industrial economist churning out forecasts in a commercial environment. Works just fine, as long as you pay only the attention to the holy doctrine of econometrics that it deserves (i.e. as little as necessary and as much as is needed) and more to your model specifications to reflect real-world behavior. It's not that hard, but then again, it's not that easy either. Then again, it's not macroeconomics, nor is it microeconomics (strictly speaking), but in-between (150 sectors covering the supply side of the economy, and no, we're not supply siders).

Key is understanding that academic work is symmetrical and tidy; the real world is asymmetrical and very messy indeed.

Eric M. Jones.

Here's my "how to run a biz" recommendation:

1) Skip the MBA.
2) Read "Up the Organization" by Robert Townsend.
3) For the small business: Read the original "The Lazy Man's Way to Riches" Skip the DYNA/PSYC crap and go straight to the mail-order business part...which is mostly obsolete, but it applies to the Internet too. It describes the nuts and bolts of running a small profitable business. Don't bother with all the rip-offs by other authors; only the original Joe Karbo edition will do.


If you apply textbook models to real world, you will be a victim of Ludic Fallacy.

If you bring practical knowledge to academics, you will be fooled by randomness.

Make a choice!


There's a simple reason for this. Most people in business are not trained in economics.

I work in the world of digital media ad sales. There is a bizarre belief among many in this industry that price is what drives overall income, rather than maximizing overall revenue with improved sell out levels. As a result, there is a striking differentiation in price between the inventory which is perceived as "best" - which is sold at a very high price, and that which is considered "worst" - and is sold as "remnant inventory" at very low prices (or not sold at all, in the case of certain very specialized publishers who seek to avoid the appearance of being low scale by running cheaper clients.

In my experience, having studied economics, I cared little about pushing price until I'd first maximized revenue by utilizing inventory effectively. At every company where I've had the opportunity to apply my approach, there have been immediate gains, which exceeded the average growth rate of the industry as a whole. Over time, as my policies continued to be applied, the gains continued to outstrip the industry rates.

I recognize there is a point at which this would stop being the case. If every site applied these rules, then each site would earn exactly (or somewhere close) to what the industry increase is, on average.

But one value of having my training is recognizing not everyone will apply these rules AND even if they do - not everyone will apply them effectively.

When I was trying to justify buying an inventory management system, a finance director asked "how much more money will it make us?" In his attempt to justify the expenditure, I explained it would not make us more money, per se, but give us the OPPORTUNITY to make more money. He failed to understand this simple fact. He took it another way.

"What if every site uses this inventory management tool? Then we lose that opportunity to make more money and everyone has the same opportunity." No, I replied. You can give me a Stradivarius, but I do not how to make it sound good. Similarly, I told him, this tool will be useful in the hands of someone who knows how to manage inventory effectively - but will be lost on someone who does not.

This disconnect will always exist, and I am glad it does. It keeps the market healthy. I am always surprised how little people in business understand about economics, however. It's amazing how people who can believe in evolution in the natural world seek to extend control over something they cannot. The concept of emergent order is so basic to an economist's understanding of life and opportunity, it makes them generally more tolerant of change and gives them more opportunities over time to be successful.


Anna D.

I agree that the disconnect between actual businesses and business as studied by economists is due to the fact that many executives do not have a strong background in economics. Economic models as a rule simplify what would be rather complex market interactions, which may give the impression that they are useless in a real world situation. This simplification is important because holding certain factors constant enables one to isolate one important cause and effect relationship. Tailoring these models to a specific company or market would be extremely beneficial if done correctly. It is true that the use of past data generally means these models give good hindsight, but that does not mean that they are not good in predicting future trends, as all companies/markets are subject to historically repetitive patterns of consumption (i.e. the business cycle). Consider those business people that do have an economics background (Donald Trump, Warren Buffet, Sam Walton, and several presidents) who have all achieved an extraordinary level of success.


john f olsen

I'd like to know why CEOs are paid such large salaries, relative to the average worker. The common explanation is a political one: CEOs are too cozy with the BOD. But perhaps there are other forces at work here? Why don't shareholders complain when top management salaries eat up significant percentages of the company's profits? Study this...