Thursday, April 09, 2015

Brand names versus reality

Some people like a shirt if it's a good shirt. Some people are obsessed with the brand name on the shirt. To some extent, this could be rational: I know nothing about cars, but I've had good experiences with cars by Toyota in the past, so I have a bias in favour of cars by Toyota. And yet, why is it that in some places and some times, brand names matter more? The simplest idea seems to be one of exposure. If the stakes are very small, I'll go by a brand name, but if not, then it makes sense to see through the brand name to the underlying reality.

There is some evidence that consumers are more brand conscious in some places than in others. The Harris Poll, 2011, says that the following proportions of consumers believe that good brands translate to quality products (from Table 5):

CountryFraction that's brand conscious
Great Britain48

Similar problems are found in the academic version of the pursuit of brand names: connections with the top universities, and publications in the top journals. Research ought to be about following your curiosity, pursuing important questions, getting novel and persuasive answers, and doing research that matters. As I wrote in this post on Indian economics, the process of recruiting and promoting researchers in India has become centred on the filtering by North American editors and referees. This chase for external brand names is exerting a corrosive effect upon the Indian academic profession. Managers of research have absolved themselves of their responsibility to judge who is a good researcher. At too many places, it's turned into a stultifying chase for brand names.

A variation of the brand name problem is the `great man syndrome'. A person scores wins in field X, and starts talking about field Y, and is able to command credibility on field Y even though the actual knowledge on that field is low.

A person is a set of brand names (where have I studied, what organisations I have worked in, what journals I have published in), a set of capabilities (character, values, ethics, knowledge) and a set of outcomes (what have I done in life). While it appears obvious that the capabilities and outcomes should matter more, some people care more about the brand names. Why is it that in some places and some times, brand names matter more?

In an ideal world, we start at a young person of age 21 and we know little about the things that matter -- emotional endurance, values, ethics, knowledge, character. So we judge the person by the brand names: "She is an NTS scholar so she must be very smart".  (I am showing my age; I believe KVPY is now the most elite club in India). As the person grows up, we can increasingly switch gears from the brand names to the person. What matters in the 20s is the brand names; in the 30s it's personality, and after that it's character.

Why might brand names matter disproportionately in India?

Let's use the notation B for the brand name, D for the data that we observe, j for our judgement about a person. We start with a very flat prior P(j); we know very little. Now we get the minimal information packet -- the brand name. Under Bayesian learning, we should P(j | B) = P(B | j) P(j) / P(B). Here we seem to do a lot of learning; if we were good Bayesians, P(B|j)/P(B) is a big number. And then we observe facts about the person D. We update P(j | D) = P(D | j) P(j) / P(D). Here, we seem to do little learning; if we were good Bayesians, P(D|j)/P(D) seems to be a small number.

What is going on? Some conjectures follow:

  1. One could say: "This has nothing to do with India; this is just `confirmation bias', a well known bug in human decision making. We are not rational Bayesian updaters, we overweight the prior and do not attach enough weight to the data. Get used to it, this problem is everywhere."  There is something to this argument. However, there is something going on e.g. people in China seems to care more about the brand name on clothes; people in India seem to care more about the brand names on the resume. Maybe humans are the same everywhere, but are there some features of the stochastic environment which make things different across space and time?
  2. In the West, the environment is stable. Trend GDP growth is 2.5%, which yields a doubling every 27 years. From age 20 to age 60, a person experiences a change from 100 to 268. This is a relaxed pace of change where people get satisfied with doing a few small steps, and can comprehend what is going on. In India, we are in an environment of far more hectic change. Trend GDP growth is 6.5%, which yields a doubling every 11 years. From age 20 to age 60, a person experiences a change from 100 to 1241. This an environment of big decisions and big consequences; this is not a comfortable locale for the cautious climber of career ladders. It is far more difficult to figure out what is going on. This is a very noisy environment. Could this high volatility generate greater conservatism i.e. inadequate updating? E.g. in money management, great returns can be owing to dumb luck, and this can happen more in a high volatility environment. In a high volatility environment, a money manager who generated high returns is less likely to have intrinsic skill -- data about performance is less informative.
  3. Inequality of knowledge could also be an issue. If I know nothing about cars, I will just fall back on brand names. When journalists know less, they will fall back on brand names -- they will think that the IIT guy must be right. If the seniors in decision making roles (who judge young people for promotions or appointments) know very little, there is a greater temptation to fall back on brand names and hire the IIT guy. Critical thinking on the part of person i about person j requires a low gap in knowledge between the two. A greater use of brand names may be inevitable in an environment of high inequality of knowledge. By this logic, the use of brand names in the world of business should be lower as the results (profit, measured in rupees) are visible for all to see.
  4. Principal-Agent problems are at work. Nobody ever got fired for hiring the IIT guy. When faced with the prospect of failure, the Agent seeks deniability by purchasing the brand name.

What goes wrong in a brand-centric world

The IIT guy may feel he has arrived. He may work less hard. He may take less risk. He doesn't have to score wins; he just has to be good enough and make it into his next job.

In academics, the research trajectory of myriad researchers is distorted by chasing brand names. A lot of people would use their lives much better if only they would dig in and research reality. This misdirection of effort results in waste. Similar things can be said all across the labour market, but it's particularly bad in academics. In the non-academic part of the world, the brand names fade away more rapidly as the person grows up.

In the US, it seems that the price paid for a brand name education is hard to justify in terms of the improvements that flow in a causal sense from that education.

One bizarre thing that I often see is an exaggerated cynicism. It's claimed that we're all clueless and ignorant and wrong. This is elaborately packaged as humility -- let's be careful to not think that thinking helps. The hidden subtext is: "Thinking is pointless, so let's just leave it to the IIT guy". By deprecating logic, we hand it over to the brand name.

    Finding underpriced assets

    The pervasive obsession with brand names has left undiscovered assets for me. I take effort to see the person rather than the brand name, and find hidden geniuses who are shunned by a brand-conscious establishment. Many heroes of the Macro/Finance Group at NIPFP fit this description. In this `security selection' process, it is relatively easy to shrug aside the brand names, but it is harder to look beyond personality and peer into character.

    I find some of the most impressive people in leadership roles in India are those who got there without brand names. This may be similar to what's being conjectured about women CEOs: It is so hard for a woman to become a CEO, she's got to be really good.


    1. Interesting article, and blog post. I agree with the pitfalls of a brand-centric world that are highlighted. Another (far-fetched) possibility is that it's a bad equilibrium in academics and industries and careers - even though someone knows they are biased they go ahead with it because others are biased too. A seller pushes brand A simply because consumers have a bias for it, even though the seller knows brand B is a much better proposition in terms of net value from a consumer's perspective.

      The more interesting feature is consumer goods and durables- clothes and accessories. I have seen from anecdotal experiences that young people (20-45) in urban India, and young Chinese people are much more brand conscious than their counterparts in the US. I have always associated this with issues of social status concerns and economic aspirations of the nouveau riche (more appropriately, 'upwardly mobile') in the developing countries experiencing rising income levels, especially in the context of India. A potentially interesting question emerges: are these people selectively brand conscious across goods? Do brands matter more in the case of products visible to the world (smartphones, clothes, cars, watches, shoes) as opposed to goods with less visibility (cookware, furniture, bedding materials....)?

    2. This is the most reassuring piece of writing i have read recently. It restored my faith in myself. This is the happiest i have felt in many days. I am just 22. It is especially relevant to me, as I have been in a supposedly premier B-School for the last 10 months and i already feel drained and sad. All that i could feel was the heat of the rat race and the madness for brands to be listed on resumes.There was no hint of originality or pure raw ambition. Your article reminds me of a poem from my school days,"Psalm of Life",by H.W. Longfellow

      Footprints, that perhaps another,
      Sailing o'er life's solemn main,
      A forlorn and shipwrecked brother,
      Seeing, shall take heart again.

      Take a bow, Sir!! Thanks a lot!

    3. See


    Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticising me is perfectly okay; uncivilised language is not. I delete any comment which is spam, has personal attacks against anyone, or uses foul language. I delete any comment which does not contribute to the intellectual discussion about the blog article in question.

    Please note: LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.