Author Archives: Michael Griffiths

Game Theory begs to be applied to the scenario described in a Knowledge@Wharton article “A Seasonal Sale Shift.”

The fourth paragraph opens like this:

With retailers pushing sales earlier and earlier and consumers waiting later and later to buy, Baker Retailing Initiative managing director Erin Armendinger compares the situation to the children’s game of chicken.

Chicken is, as we all know, a special application of the classic Prisoner’s Dilemma game – where the mutually agreeable situation is unstable, because both sides benefit from departing from it (as long as the other does not).

Not to mention the massive coordination failure between competing retailers: if a competitor starts offering markdowns, you have to as well or you’re screwed. Oh, the Betrand model, how easily you screw over retailers.

It’s a situation that begs for a game-theoretic analysis, but the article doesn’t even mention the phrase.

Also, they misuse the term “value proposition.” They’re conflating the idea of “value” (dropped wholesale by the profession of economics because it was too hard to pin down). The value proposition encapsulates the experience the customer has due to your product – rather similar to the famous Drucker quote “the world wants holes, not drill bits.” Price is always an aspect – especially in retail – but it’s not really related to the value proposition as such.

Now, retailers actually are changing their value proposition. From the article:

For example, apparel retailers have been changing their supply chains and inventory as part of what’s known as a "wear now" strategy. While in the past, suggests Armendinger, retailers rolled out merchandise for a new season on a particular day — changing to displays of sweaters and corduroy pants in the height of summer’s heat — many are now offering a balance of clothing so people shopping in August can still find shorts and short-sleeved shirts.

Funny how that works.

Knowledge@Wharton recently featured an article explaining why ranking employees relative to their peers can backfire.

The astonishing thing is how rapidly people have forgotten two millennia of leadership.

The “astonishing” finding in the article – that  is, astonishing to people who have managed to avoid 40 years of work in employee motivation by psychologists – is that (i) giving employees feedback comparing them to their peers can cause resentment, and (ii) financial rewards do not always correlate with higher attainment.

To look at the the issue another way: why was this astonishing? What’s necessary to believe such that people believed that doing those two things would actually improve employee productivity?

Simple: that people (employees) are highly rational actors knowingly engaged in competition with their peers for a finite amount of resources (compensation) who attempt to optimize their personal income with no regard to others.

That doesn’t sound terribly human – or terribly friendly, for that matter – so why did people act as if it were true?

Such is the influence of economics, the magical discipline which is two parts mathematical rigor and one part magical thinking (which directs the other two parts).

For purely practical reasons, early economists constructed models explanations of the economy that assumed that actors were perfectly rational. It was sort of a mind game: What would the perfect economy look like? To what extent is our current economy perfect?

It was a practical attempt because data simply wasn’t available to run empirical studies on something like the macroeconomy, or even large markets. The computational power simply wasn’t available to create sophisticated models using that could account for non-rational behavior.

Why? Because economists needed – that is, they weren’t capable of developing anything else – a model that was deterministic. An individual, placed in the same situation, would make the same decisions (assuming their situation has not changed). A non-deterministic model is both far more difficult to create, requires far more computation, and is also considerably less useful.

The second reasons is simpler – economists weren’t trying to predict what an individual would do, they were trying to predict what all individuals would do. That is, they assumed that the average of all actions in a market would be rational. And they found a reasonable amount of evidence supporting that, where they expected to (in markets that had other assumptions of competition).

This explanation is vastly oversimplified and ignores both some thinkers and some developments.

In any event, like all academia, a popular idea remains a popular idea – until the tides of reality can o’ercome it. The rationalism of individuals was repeatedly challenged (frequently by economists), but economists could successfully point out that they were talking to the average of all actions, so as long as the average turned out to be roughly rational, they were OK. Since economists never drifted down to individuals, everything worked out dandy.

The problem came when people – and I’m looking at business – who did deal with individuals attempted to apply the lessons of economics to running their business. After all, economics began as a normative science to figure out what to do – so applying those normative guidelines should lead to greater efficiency.

Except that the average of all actions is not the same as the average action. It’s the same as the myth of the perfectly normal man – he doesn’t exist. It’s a statistical accident, really, that arises when you take a distribution and reduce it to a single number.

Rationalism for individuals had all the advantages for business (HR and all) that it had for economists: it allows for a (mostly) deterministic system. Given a scenario, the rational person will make the rational choice. No more fuzzy-duddy special exceptions for snowflake-special individuals: just cold, hard, unfeeling, highly efficient rules.

And thus died the two-millennia history of leadership rooted in the idea of inspiration; or making people fight for a cause greater than themselves, producing something that stretched beyond their own narrow lives.

The death of the dream to change the world – replaced, as it were, by a mission statement to align interests. Oh, and a vision statement so you know where you’re going, allowing you to plan how best to get there.

On a completely unrelated note, the study the Knowledge@Wharton article references is completely inapplicable to the real world. It is, however, backed by a mountain of empirical and theoretical psychology.

The Republican party seems to be pushing the perennial myth that the “Free Market” will make the healthcare system more efficient. This ignores a wealth of evidence by respected neoclassical economists (such as Kenneth Arrow), but also ignores plain Micro 101.

To clarify, a few things are needed for a efficient market:

1. Consumers have an indifference curve for consumption.
2. Suppliers are price-takers; they do not discriminate on price.
3. Consumers have full information about all products on the market.
4. There are no transaction costs and no barriers to entry/exit.
5. Homogenous products.
6. Constant returns to scale.

It’s possible to quibble with one or two (you can argue that one is redundant, and others add more requirements, etc) but that’s the general necessary assumptions.

Now, the free market only exerts its magic is all requirements are fulfilled. Obviously, it’s not a binary result, so if you violate an assumption to a greater or lesser degree, the market becomes less efficient – and thus less “magical.”

This is not controversial. I just want to repeat: this is taught in every Micro 101 post, and freely acknowledged by every neoclassical economist.

Let’s consider the healthcare market.

First of all, consumers do not have set indifference curves for healthcare consumption. Certainly, there are exceptions; and there are some healthcare providers where people do have such indifference curves (non-essential providers).

However, the important stuff is the unpredictable stuff. It’s worth nothing that medical expenses are involved in about half of bankruptcies.

But from a logical standpoint, suddenly discovering cancer changes your desire to consume healthcare. This is not by choice. It’s in response to changing events.

As a consequence, forcing everyone to pay out of pocket for medical expenses (taking it to the extreme) would do nothing but forced people who have unfortunate medical problems to suffer and die – through no fault of their own, and outside of their ability to control.

This is the “moral issue” that Kenneth Arrow pointed out: do we, as a society, want to let citizens of our democracy sicken and die simply because they have the bad luck to fall ill?

Not to mention the related issues – healthcare is not homogenous (every person requires different treatment), suppliers are not price-takers, consumers never have full information on treatment options and costs, and there are tremendous transaction costs to undergo a procedure.

Any of those would be enough to kill the possibility of an efficient healthcare market.

But the deeper problem is assuming that people can ration their healthcare consumption, according to how healthy they want to be, and that costs for each person will be roughly equivalent (varying only in the amount they choose to consume).

That’s a fantasy, and a piss poor one at that.

Of course, healthcare has a number of problems currently – one of the major ones it the lack of decent incentives. How many people die in hospitals for lack of basic hygiene? For medical mistakes, like hooking up a feeding tube to a blood vein? And how is it that a hospital that tries to reduce medical accidents actually ends up losing money – because the real profit is in treating more serious disorders, so it pays to make someone sicker before you make them better?

Those problems can be solved either through incentive alteration (changing the entire pay system of health insurance) or through regulation; for obvious reasons, changing and enforcing regulation is the easier (politically feasible, economically feasible, and successful) way to fix or at least alleviate those problems.

Making healthcare more of a free market will not.

However much people talk of “moral hazard” and “tort reform.”