If you’re taught microeconomics in India, teachers will offer bargaining for an auto-rickshaw or buying tea at a chai kadai, as examples of price discrimination. Dress well or hail from a wealthy area, and auto-wallahs are bound to charge you more. All of this may be true, but I have another suggestion that is particularly relevant to cash-intensive economies across Asia. I’m writing this post because I suspect it has rather deep welfare effects.

Chennaiites (who are not blessed with meter-using autos) employ handy heuristics to anchor a negotiation. Maybe “10 rupees a kilometer”. In Bangalore at night it might be “meter plus 50”. Such guesses and tactics – though ephemeral with punishing inflation – underpin many of India’s informal transactions.

But a cash economy lends itself to a price discrimination that requires no thought or design. The market rate for a ride from Adyar to T. Nagar might be 36 rupees. A good number of passengers will have exactly this in cash. It is less likely, though, that wealthier Indians will have anything smaller than a Rs. 50 note. The autowallah will generally have “no change” and hence reap Rs. 14 in supernormal profit.

Similarly, I find myself in a pickle purchasing a Rs. 6 chai with the relatively large notes in my pocket. If I’m lucky, I can pay Rs. 50, and receive four ten Rupee notes in change, loosing Rs. 4 in surplus.

Another mechanism is at play. This post from Tyler Cowen suggests that in a salary negotiation, people who are able to offer precise prices (like $57,500) implicitly convey a deep knowledge of that labor market, increasing their bargaining power. If I carry only larger notes (certainly no paisa coins!), I’m limited in offering only round prices which conveys both a) a possible lack of market knowledge, which would be true and b) a lack of small bills, suggesting a higher reservation price.

Auto-rickshaws, chai kadais, and other informal venues of commerce are still widely used by middle and upper class Indians. The market equilibrium for imperfect conditions would hence otherwise exclude poorer customers (that is the price exceeds marginal cost) In India, it’s also rather difficult (even for an experienced salesman) to estimate reservation prices based solely on one’s clothes and phone. Of course, if you flaunt an iPhone, all hope may be lost, but for the interquartile range of consumers, it is far more difficult.

To this end, the granularity of denomination one carries probably plays a large role in redistribution of welfare. This has broad consequences. The Rs. 14 a driver gains because I carry nothing smaller than a 50 Rupee note is a very significant chunk of his disposable income. Furthermore, government regulation, occupational licensing, and corruption make many such informal venues of commerce monopolistic in nature.

To the extent that these are imperfect markets, frictions from a cash-intensive economy might just approximate socially optimal conditions.