...helping clients in complex environments make better decisions, faster.

CALL NOW: 404.898.9096


Archive for September, 2013

If you want Wal-Mart to have a labor force like Trader Joe’s and Costco, you probably want them to have a business model like Trader Joe’s and Costco

From Why Wal-Mart Will Never Pay Like Costco by Megan McArdle.

In other words, Trader Joe’s and Costco are the specialty grocer and warehouse club for an affluent, educated college demographic. They woo this crowd with a stripped-down array of high quality stock-keeping units, and high-quality customer service. The high wages produce the high levels of customer service, and the small number of products are what allow them to pay the high wages. Fewer products to handle (and restock) lowers the labor intensity of your operation. In the case of Trader Joe’s, it also dramatically decreases the amount of space you need for your supermarket … which in turn is why their revenue per square foot is so high. (Costco solves this problem by leaving the stuff on pallets, so that you can be your own stockboy).

Both these strategies work in part because very few people expect to do all their shopping at Trader Joe’s, and no one expects to do all their shopping at Costco. They don’t need to be comprehensive. Supermarkets, and Wal-Mart, have to devote a lot of shelf space, and labor, to products that don’t turn over that often.

Wal-Mart’s customers expect a very broad array of goods, because they’re a department store, not a specialty retailer; lots of people rely on Wal-Mart for their regular weekly shopping. The retailer has tried to cut the number of SKUs it carries, but ended up having to put them back, because it cost them in complaints, and sales. That means more labor, and lower profits per square foot. It also means that when you ask a clerk where something is, he’s likely to have no idea, because no person could master 108,000 SKUs. Even if Wal-Mart did pay a higher wage, you wouldn’t get the kind of easy, effortless service that you do at Trader Joe’s because the business models are just too different. If your business model inherently requires a lot of low-skill labor, efficiency wages don’t necessarily make financial sense.

That’s not the only reason that the Trader Joe’s/Costco model wouldn’t work for Wal-Mart. For one thing, it’s no accident that the high-wage favorites cited by activists tend to serve the affluent; lower income households can’t afford to pay extra for top-notch service. If it really matters to you whether you pay 50 cents a loaf less for generic bread, you’re not going to go to the specialty store where the organic produce is super-cheap and the clerk gave a cookie to your kid. Every time I write about Wal-Mart (or McDonald’s, or [insert store here]), several people will e-mail, or tweet, or come into the comments to say they’d be happy to pay 25 percent more for their Big Mac or their Wal-Mart goods if it means that the workers are well paid. I have taken to asking them how often they go to Wal-Mart or McDonald’s. So far, no one has reported going as often as once a week; the modal answer is a sudden disappearance from the conversation. If I had to guess, I’d estimate that most of the people making such statements go to Wal-Mart or McDonald’s only on road trips.

However, there are people for whom the McDonald’s Dollar Menu is a bit of a splurge, and Wal-Mart’s prices mean an extra pair of shoes for the kids. Those people might theoretically favor high wages, but they do not act on those beliefs — just witness last Thanksgiving’s union action against Wal-Mart, which featured indifferent crowds streaming past a handful of activists, most of whom did not actually work for Wal-Mart.

If you want Wal-Mart to have a labor force like Trader Joe’s and Costco, you probably want them to have a business model like Trader Joe’s and Costco — which is to say that you want them to have a customer demographic like Trader Joe’s and Costco. Obviously if you belong to that demographic — which is to say, if you’re a policy analyst, or a magazine writer — then this sounds like a splendid idea. To Wal-Mart’s actual customer base, however, it might sound like “take your business somewhere else.”

There is a correct answer to that question, but it’s unlikely we’ll ever know what it was.

From Why Do Education and Health Care Cost So Much? by Megan McArdle.  A great example of the challenges related to causal density.  We may accurately identify all the causes of an outcome but still not be able, because of poor understanding of the relationships between root causes, to predict outcomes.  Absent accurate prediction, we don’t really understand the nature of a problem at all.

So how do we explain health care and college cost inflation? Well, health care economist David Cutler once offered me the following observation: In health care, as in education, the output is very important, and impossible to measure accurately. Two 65-year-olds check into two hospitals with pneumonia; one lives, one dies. Was the difference in the medical care, or their constitutions, or the bacteria that infected them? There is a correct answer to that question, but it’s unlikely we’ll ever know what it was.

Similarly, two students go to different colleges; one flunks out, while the other gets a Rhodes Scholarship. Is one school better, or is one student? You can’t even answer these questions by aggregating data; better schools may attract better students. Even when you control for income and parental education, you’re left with what researchers call “omitted variable bias” — a better school may attract more motivated and education-oriented parents to enroll their kids there.

So on the one hand, we have two inelastic goods with a high perceived need; and on the other hand, you have no way to measure quality of output. The result is that we keep increasing the inputs: the expensive professors and doctors and research and facilities.

I would quibble with McArdle.  There are actually two problems.  It is true that it is hard to measure education and health outcomes and that is a challenge.  But even if we were able to measure with great precision and accuracy, that is still not the same as forecasting.  Measuring is a predicate to forecasting.

If we precisely and accurately measure our initiating action X, we want to know with some level of accuracy and certainty that X will lead to Y, the outcome we desire.  If we cannot predict the outcome, it means we don’t understand the relationship between and among the various causes.

Life is short, and Art long; the crisis fleeting; experience perilous, and decision difficult.

Aphorisms by Hippocrates, the very first aphorism:

Life is short, and Art long; the crisis fleeting; experience perilous, and decision difficult. The physician must not only be prepared to do what is right himself, but also to make the patient, the attendants, and externals cooperate.

A neat summation that remains true in virtually all fields today.  It takes a long time to develop skills (Art long) compared to the allotted time we have to develop those skills (life is short), the pace of change is fast and we only have one chance to get it right (the crisis fleeting), the consequences are great (experience perilous) and as always the “decision difficult.”  The second sentence is not perhaps as crisp but is just as consequential.  A good decision without support from all parties in the context (the patient, the attendants, and externals) is like as not to fail.  The challenge is that in the modern era, rarely can we “make” others cooperate, rather we have to convince, motivate, or incent them to cooperate, an endeavor fraught with variable outcomes.  A variability that is less and less desirable, the greater the consequences arising from the decision.

The Latin original is Vita brevis, ars longa, occasio praeceps, experimentum periculosum, iudicium difficile.

Most frequent errors, fallacies, and biases when decision-making

Looking for something that might tell me how often logical fallacies and cognitive biases occur in discussions, I could find nothing at all.  Not willing to let go, I resorted to using N-grams.  It has the drawback that some fallacies and biases are terms commonly used in other contexts (ex: false memory) or returned no results (ex: normalcy bias).  I lost about hundred biases and fallacies from this weakness, though generally more obscure or nuanced biases and fallacies.  This was about half the population.  Of the remaining hundred or so, I was able to obtain an N-gram number and then rank from largest to smallest.

All this tells us is the degree to which specific biases, errors and fallacies are being discussed in books.  I am making the bold inference that specific biases and fallacies which are discussed frequently are correspondingly more common or more problematic (i.e. perhaps they don’t occur that often but are more consequential when they do).  So having caveated the corpus to death, I present the top most commonly discussed biases and fallacies in a whispering ghost of a list.

These would seem to be the errors, fallacies, and biases you are most likely to encounter when working with a group to reach an empirical, logical, and evidence based decision.

Anecdotal Evidence
False assumptions
Cognitive Dissonance
Fallacy of composition
Unstated Assumptions
Slippery Slope
Selective perception
Halo effect
Argumentum Ad hominem
Illusory correlation
Source Credibility
Forer effect (aka Barnum effect)
Sunk cost bias
Begging the Question
Fundamental attribution error
Generalizing personalities
Hindsight bias

Not quite the list or order I would have expected, but not completely out of the realm of probability.  The top ten in particular are broadly consistent with my experience in terms of mistakes teams make when trying to arrive at decisions.

Solution to low public transportation utilization

From Mobility for the Poor: Car-Sharing, Car Loans, and the Limits of Public Transit by Jeff Khau.  An example of the importance of establishing the difference between correlation and causation; of the importance of directionality of causation; of context; of root cause analysis; and goal definition.

Theoretically, one can look at this graph and legitimately make the argument that in order to increase public utilization of public transportation, one ought to increase the average commute time.  It is a good exercise in critical thinking to spot the fallacy of such an interpretation.

Confirmation Bias, Illustrated

From Six Chix


Precision and Accuracy

While it might seem an exercise in splitting hairs, in Approximate quotations by Mark Liberman, the author opens up the basis for a good discussion on the differences between approximation, precision and accuracy.  All three aspects are important and necessary but they are different and are each more appropriate in different contexts.

An approximate quote can yield a better general sense of what was being communicated but then the accuracy depends on the interpretation of the journalist.  A direct quote as from a transcript is more precise but more burdensome to the reader.

Sometimes the need for precision is paramount.  On other occasions, an approximation is more efficient.  It is a matter of horses for courses, as long as we keep the distinction between precision and accuracy clear.

Fairy tales masquerading as evidence

Science bible stories, take 27 by Mark Liberman is a useful discussion (including in the comments) about the tendency of media to take up a topical research paper without regard to the methodological robustness of the study or whether the results are meaningfully true.

As I observed a few years ago, “scientific studies”  have taken over the place that bible stories used to occupy. It’s only fundamentalists like me who worry about whether they’re true. For most people, it’s enough that they can be interpreted to be morally instructive.


I’d add a third important factor: by and large, the “wise men” (and now the “wise women”) don’t really care about whether the empirical and theoretical foundations of their opinions are sound . They care about readers, ratings, and reputation — and in some cases about political outcomes or cultural values —  with truth relevant only insofar as it affects those goals.

I think Liberman is correct.  People rarely consider what evidence they need in order to make an argument, instead they go after information that is convenient to get.  At the same time, the market structure for ideas and information is such that there are incentives to produce affirming information to a range of prejudices, regardless of the truth of the matter.  Elsewhere I refer to this as cognitive pollution as it constitutes dirt in the system that tends to occlude rather than clarify.

RELATED:  The culturomic psychology of urbanization by Mark Liberman

To analyze does not necessarily mean to produce useful information

From Reviewing the Movies: Audiences vs. Critics by Catherine Rampell.

It is a fair and interesting question or set of questions.  Do audiences and critics assess movies in different ways?  If so, in what ways do they differ?  Which views, audience or critics, provide a better forecast of future performance?  These questions apply to art, sports, books, etc. There are answers to some of these questions.  The general informed public and specialists do tend to review things differently.  General informed public tend to factor in more context and larger macro considerations than do specialists.  General informed public tend to be better forecasters than are specialists.  Nate Silver covers a lot of this in his The Signal and The Noise: Why So Many Predictions Fail – but Some Don’t
What is notable is that Rampell asks a legitimate question and has an idea on how to answer the question.  Her error is to use information that is available (Rotten Tomatoes Database) rather than information that is needed.  There is a fairly detailed critique of her analysis in the comments.  The article serves as an example of Selection Bias (the distortion of a statistical analysis, resulting from the method of collecting samples) and Information Bias (the tendency to seek information even when it cannot affect action), and possible Mere Exposure Bias (the tendency to express undue liking for things merely because of familiarity with them).


Almost all Americans devoutly believe, the liberal, market principles on which our country is built will triumph around the world.

From Bambi Meets Godzilla In The Middle East by Walter Russell Mead.  Read the whole thing.

The end of history, which AI founder Francis Fukuyama used to describe the historical implications of the Cold War, is to American political philosophy what the Second Coming is to Christians. In the end, almost all Americans devoutly believe, the liberal, market principles on which our country is built will triumph around the world. Asia, Africa, South America, the Middle East and even Russia will some day become democratic societies with market economies softened by welfare states and social safety nets. As a nation, we believe that democracy is both morally better and more practical than other forms of government, and that a regulated market economy offers the only long term path to national prosperity. As democracy and capitalism spread their wonder-working wings across the world, peace will descend on suffering humanity and history as we’ve known it will be at an end.


It seems misanthropic to doubt that a particular country isn’t on the road to freedom and prosperity, and it also seems like heresy against our national creed. That tendency is reinforced among our policy elite and chattering classes. The “experts” ought to know better and be more skeptical, but they are often more naive and more dogmatic than the American people at large. It is often the best educated and connected who are most confident, for example, that political science maxims work better than historical knowledge and reflection when it comes to analyzing events and predicting developments. When democratic peace theory or some other beautiful intellectual system (backed with regressions and statistically significant correlations in all their austere beauty) adds its weight to the national political religion, a reasonable faith can morph into blind zeal. Bad things often follow.

What Americans often miss is that while democratic liberal capitalism may be where humanity is heading, not everybody is going to get there tomorrow. This is not simply because some leaders selfishly seek their own power or because evil ideologies take root in unhappy lands. It is also because while liberal capitalist democracy may well be the best way to order human societies from an abstract point of view, not every human society is ready and able to do walk that road now. Some aren’t ready because like Haiti they face such crippling problems that having a government, any government, that effectively enforces the law and provides basic services across the country is beyond their grasp. Some aren’t ready because religious or ethnic tensions would rip a particular country apart and cause civil war. Some aren’t ready because the gap between the values, social structures and culture of a particular society make various aspects of liberal capitalism either distasteful or impractical. In many places, the fact that liberal democratic capitalism is historically associated with western imperialism and arrogance has poisoned the well. People simply do not believe that this foreign system will work for them, and they blame many of the problems they face on the countries in Europe and North America who so loudly proclaim the superiority of a system they feel has victimized them.


Americans need to face an unpleasant fact: while American values may be the answer long term to the Middle East’s problems, they are largely irrelevant to much that is happening there now. We are not going to stop terrorism, at least not in the short or middle term, by building prosperous democratic societies in the Middle East. We can’t fix Pakistan, we can’t fix Egypt, we can’t fix Iraq, we can’t fix Saudi Arabia and we can’t fix Syria. Not even the people who live in those countries can fix them at this point; what has gone wrong is so deeply rooted and so multifaceted that nothing anybody can do will turn them into good candidates for membership in the European Union anytime soon. If we could turn Pakistan into Denmark, the terrorists there would probably settle down—but that isn’t going to happen on any policy-relevant timetable. We must deal with terrorism (and our other interests in the region) in a world in which the basic conditions that breed terrorists aren’t going away.

Decision Clarity Consulting has no affiliation with the Scheier Group, owner of the DECISION CLARITY®, and the Scheier Group does not sponsor, endorse, or control any material on this site. Decision Clarity Consulting, in accordance with an agreement with Scheier Group, performs no work with non-profit or government entities.