Wednesday, October 29, 2014

Economic theory of value, price and utility

Price is a very sensitive topic in microeconomics. There has been a lot of work done in this domain. Unfortunately most of it is completely misguided by attempts to theorize about price from a normative thinking process of ‘what should be the price?’ or ‘how should a rational individual think about price?’

While going through a lecture on Coursera on Neuroeconomics, I revisited this question again in my head. Partly it was prompted by the idea in the lecture itself – that the ‘value’ something has for an individual has been generally expressed in terms of price or utility in economic theorizing and that the Neuroeconomic approach to it is to start with neuronal firing rate in response to a good/activity and its like reward or punishment value. Interesting and promising approach indeed.

What i was tangentially thinking about was something related but different. I was wondering if the entire framework of answering the question ‘what is something worth to an individual?’ is mistaken. Here’s why.
The conventional attempts at finding out worth of something to someone is generally focused on the utility of the good/service to the person and some estimate of the value of the same. There is an implicit assumption that this is constant across space and time and individuals. All three are faulty assumptions – there are not even good first level approximations. That is the reason behind my question above.

Part of the reason is the dependence of value on location, time and individual. Part is also the very approach of trying to model it like this. When one implicitly assumes that something has an objective value and that just needs to be determined through some observation, one is already committing the folly of creating an imaginary quantity (objective value). One is then likely to fall into the traps of calling something over-valued, under-valued, over-priced and so on.

My view on value and price is as follows.
Value of something is inherently and fundamentally subjective as well as a function of time and place. It is also reflexively dependent on perception, network effects and inference about who else is a consumer.
First of this is simple to demonstrate.
  1. Dependence on person: I like QWERTY keyboard phones while someone else values bigger screen. Even at an aggregate product level, someone may like orange juice as a refresher while her friend might prefer a quick call with her fiancĂ©. Given a specific good, different people will value it very differently – I love tea, my wife does not have tea at all and i know of a lot of people who have intermediate levels of liking for it.
  2. Dependence on time: Food is valuable when hunger strikes, music adds value to a pub night, cab service is more valuable in monsoon and wee hours etc etc
  3. Dependence on place: Mineral water at the top of a mountain is more valuable than in the middle of the city, binoculars are of value in a desert but not as much in a jungle and so on.

Note that if we start without the baggage of goods having a similar price across people, time and place, the above divergence would point us in the direction of a non-unique value and price automatically. Only if we start with the state of the world as it is today that we would attempt to figure out explanations and workarounds to this obvious state of affairs.

Second set of divergences is more subtle. It is also more applicable to modern branded goods than to commodities.
  1. Dependence on network effects: An app that my friends use is more valuable than one that nobody uses.
  2. Dependence on perception: If Hyundai cars are not perceived to be premium, I would flinch in buying a feature-rich Hyundai car for a high price point.
  3. Dependence on inference about who else is a consumer: If everyone is using Ray-Ban shades, I would also join in. Sometimes this also has adverse effect – if everyone (‘the masses’) is using Gucci, i better stop using it (it has become ‘pedestrian’)

The value derived from a good/service is hence a complex function of all of the above factors. Just to be clear, these are not factors that are small deviations around a secular level. This is precisely the mistaken stance conventional microeconomics takes. Most economists would acknowledge the presence of these deviations. However, in the name of tractability and approximations, they would make an undefendable leap of faith that a large proportion of value is independent of this and thus can be thought of as objective.

The other angle often ignored in classical economic theory with regard to price is how the producers determine it. Most often, the over-generalized response of economists is that perfect competition persists in most cases and the producers charge a price that is equal to marginal cost of production. This is again normative. It is demonstrated in real life only in a small minority of cases where a highly uniform commodity is traded in a highly transparent manner (crude oil, steel etc). For most real life cases, price is nearly arbitrarily determined – producers do take into account the cost plus logic but more often than not, the linkage is reflexive. If something fetches good price, its input goods start to reflect that as well through higher pull. It is not only that input good become costlier and thus output good catch up in terms of price.

Just like we don’t know how each specific biological species began its journey on earth, the pricing history of each specific good and service is hard to trace back. Since everything has something or other as input (including labor) which has its own price, it is hard to study the absolute level of price of anything in isolation. However, that should not make us complacent about the origin of price-levels.

A feedback loop exists between consumers and producers as well and it is not simply a matter of a producer looking to offer at a price above a certain minimum and a consumer making sure of a bidding war each time she is looking to buy something. Real life transactions have a lot of influence of behavioural factors as well as institutional factors. Some prices are simply a matter of habit, others of arbitrary anchors and so on. On this base case the consumer producer feedback takes place.

In summary, value is not identifiable in an objective sense. Hence a uniform price for a good/service is an arbitrary imposition of simplicity. I think it is not hard to think about constantly varying prices of goods and services across people, place and time. The efficient market enthusiasts will jump at this suggestion and cry ‘arbitrage’. However, insofar as consumer goods are concerned, it is hard to imagine majority of people engaging in an arbitrage about making someone else buy something or hoarding some stuff because it is cheap at that time.

In some cases this already happens – but it is limited to dependence on place. Convenience stores sell things at a major premium to supermarkets. However, time dependence, situational factors and person dependence is almost never factored in. Even more so, the perception effects, network effects and so on are rarely if ever incorporated into pricing. Even the limited space dependence of the type of convenience stores vs supermarkets is a matter of practice – not entirely explained in economic theory.

No comments: