between the buyer and the seller – Pertinent Observations

Ratings revisited

Sometimes I get a bit narcissistic, and check how my book is doing. I log on to the seller portal to see how many copies have been sold. I go to the Amazon page and see what are the other books that people who have bought my book are buying (on the US store it’s Ray Dalio’s Principles, as of now. On the UK and India stores, Sidin’s Bombay Fever is the beer to my book’s diapers).

And then I check if there are new reviews of my book. When friends write them, they notify me, so it’s easy to track. What I discover when I visit my Amazon page are the reviews written by people I don’t know. And so far, most of them have been good.

So today was one of those narcissistic days, and I was initially a bit disappointed to see a new four-star review. I started wondering what this person found wrong with my book. And then I read through the review and found it to be wholly positive.

A quick conversation with the wife followed, and she pointed out that this reviewer perhaps reserves five stars for the exceptional. And then my mind went back to this topic that I’d blogged about way back in 2015 – about rating systems.

The “4.8” score that Amazon gives as an average of all the ratings on my book so far is a rather crude measure – since one reviewer’s 4* rating might differ significantly from another reviewer’s.

For example, my “default rating” for a book might be 5/5, with 4/5 reserved for books I don’t like and 3/5 for atrocious books. On the other hand, you might use the “full scale” and use 3/5 as your average rating, giving 4 for books you really like and very rarely giving a 5.

By simply taking an arithmetic average of ratings, it is possible to overstate the quality of a product that has for whatever reason been rated mostly by people with high default ratings (such a correlation is plausible). Similarly a low average rating for a product might mask the fact that it was rated by people who inherently give low ratings.

As I argue in the penultimate chapter of my book (or maybe the chapter before that – it’s been a while since I finished it), one way that platforms foster transactions is by increasing information flow between the buyer and the seller (this is one thing I’ve gotten good at – plugging my book’s name in random sentences), and one way to do this is by sharing reviews and ratings.

From this perspective, for a platform’s judgment on a product or seller (usually it’s the seller, but for products such as AirBnb, information about buyers also matters) to be credible, it is important that they be aggregated in the right manner.

One way to do this is to use some kind of a Z-score (relative to other ratings that the rater has given) and then come up with a normalised rating. But then this needs to be readjusted for the quality of the other items that this rater has rated. So you can think of some kind of a Singular Value Decomposition you can perform on ratings to find out the “true value” of a product (ok this is an achievement – using a linear algebra reference given how badly I suck in the topic).

I mean – it need not be THAT complicated, but the basic point is that it is important that platforms aggregate ratings in the right manner in order to convey accurate information about counterparties.

Thaler and Uber and surge pricing

I’m writing about Uber after a really long time on this blog. Basically I’d gotten tired of writing about the company and its ideas, and once I wrote a chapter about dynamic pricing in cabs in my book, there was simply nothing more to say.

Now, the Nobel Prize to Richard Thaler and his comments sometime back about Uber’s surge pricing has given me reason to revisit this topic, though I’ll keep it short.

Basically Thaler makes the point that when businesses are greedy and seen to be gouging customers in times of high demand, they might lose future demand from the same customers. In his 2015 book Misbehaving (which I borrowed from the local library a few months ago but never got down to reading), he talks specifically about Uber, and about how price gouging isn’t a great idea.

This has been reported across both mainstream and social media over the last couple of days as if Thaler is completely against the concept of surge pricing itself. For example, in this piece about Thaler, Pramit Bhattacharya of Mint introduces the concept of surge pricing and says:

Thaler was an early critic of this model. In his 2015 book Misbehaving: The Making of Behavioral Economics, Thaler argues that temporary spikes in demand, “from blizzards to rock star deaths, are an especially bad time for any business to appear greedy”. He argues that to build long-term relationships with customers, firms must be seen as “fair” and not just efficient, and that this often involves giving up on short-term profits even if customers may be willing to pay more at that point to avail themselves of its product or service.

At first sight, it is puzzling that an economist would be against the principle of dynamic pricing, since it helps the marketplace allocate resources more effectively and more importantly, use price as an information mechanism to massively improve liquidity in the system. But Thaler’s views on the topic are more nuanced. To continue to quote from Pramit’s piece:

“I love Uber as a service,” writes Thaler. “But if I were their consultant, or a shareholder, I would suggest that they simply cap surges to something like a multiple of three times the usual fare. You might wonder where the number three came from. That is my vague impression of the range of prices that one normally sees for products such as hotel rooms and plane tickets that have prices dependent on supply and demand. Furthermore, these services sell out at the most popular times, meaning that the owners are intentionally setting the prices too low during the peak season.

Thaler is NOT suggesting that Uber not use dynamic pricing – the information and liquidity effects of that are too massive to compensate for occasionally pissing off passengers. What he suggests, however, is that the surge be CAPPED, perhaps at a multiple of three.

There is a point after which dynamic pricing ceases to serve any value in terms of information and liquidity, and its sole purpose is to ensure efficient allocation of resources at that particular instant in time. At such levels, though, the cost of pissing off customers is also rather high. And Thaler suggests that 3 is the multiple at which the benefits of allocation start getting weighed down by the costs of pissing off passengers.

This is exactly what I’ve been proposing in terms of cab regulation for a couple of years now, though I don’t think I’ve put it down in writing anywhere. That rather than banning these services from not using dynamic pricing at all, a second best solution for a regulator who wants to prevent “price gouging” is to have a fare cap, and to set the cap high enough that there is enough room for the marketplaces to manoeuvre and use price as a mechanism to exchange information and boost liquidity.

Also, the price cap should be set in a way that marketplaces have flexibility in how they will arrive at the final price as long as it is within the cap – regulators might say that the total fare may not exceed a certain multiple of the distance and time or whatever, but they should not dictate how the marketplace precisely arrives at the price – since calculation of transaction cost in taxi pricing has historically been a hard problem and one of the main ways in which marketplaces such as Uber bring efficiency is in solving this problem in an innovative manner using technology.

For more on this topic, listen to my podcast with Amit Varma about how taxi marketplaces such as Uber use surge pricing to improve liquidity.

For even more on the topic, read my book Between the buyer and the seller which has a long chapter dedicated to the topic,

Book Release

So my book Between the buyer and the seller is now available on Amazon, in both print and kindle versions. You can go here to buy. Thanks to Amazon’s print on demand service, it’s available worldwide.

It’s been a long time coming. I completed the first draft way back in April 2016. Writing it was no easy task, but was definitely helped by the presence of one awesome coffee shop close to where I was staying in Barcelona.

Having written one draft, I went around finding publishers. It wasn’t a trivial process. In the process, I found out enough about the publishing industry to get a new prologue for the book (I guess that should be part of the Kindle sample).

And then in the course of the backs and forths with the publishers I found a lot of what I’d written to be absolute shit, and so revised the book two times. Then in December last year, the Takshashila Institution decided to publish it.

And then they sent it to some experts for expert opinion. Said opinion came back positive but with some suggestions. So I revised the book yet another time and implemented these suggestions. Then there was the copy editing process and yet another revision. Then the book design (if not anything, doesn’t it at least look good?) and typesetting and stuff. And formatting.

In the meantime, Shashi Tharoor and Bibek Debroy wrote some nice blurbs for the book – they’re printed at the back of the book now. And then some more hoops and procedures and printing and publishing and fighting with Amazon and the book is now out! For you to purchase.

I want to put out a special word of thanks to Anupam Manur, who has effectively “produced” the book, managing the entire process on Takshashila’s behalf. He’s been patient with my periodic abuses, and diligently got work done. The night before his wedding, he was up fixing some stuff on Amazon for the book.

Anyway, enough of my story. Now go buy the book and read it. Let me, and others, know what you think of the book. And spread the word!

Oh, and I want to thank all of you, my patient blog readers, for the encouragement through the last 13 years. It’s your collective effort and support that has made me a better writer, and resulted in this book coming out!

Platform as a platform

This afternoon, as I was getting off the tube, I looked at the railway platform, and wondered how it compared to “platforms” as we now know in the context of “platform economics“. For those of you under a rock, platform economics talks about the economics of “platforms” that bring together two sides of a market to interact.

In that sense, Uber is a platform connecting drivers to passengers. Ebay is a platform connecting buyers and sellers of used goods. Paypal is a platform connecting people who want to pay and those who want to receive payment. And so forth (these are all textbook examples nowadays).

So is the railway platform a platform? And if not, is it correct that we refer to entities that run two-sided markets as platforms (arguably, the most intuitive meaning of the word “platform” in the last hundred or so years has been in the railway context)? These were some of the questions I grappled with as I walked along the length of the platform at Ealing Broadway.

For those of you who’re not in the know, I’ve written a book on market design. The Takshashila Institution is publishing it, and the book should be out fairly soon (manuscript is complete, but there’s still plenty to do). In that book, I have a chapter on taxi marketplaces such as Uber/Lyft/Ola, and how they’ve transformed the efficiency of the taxi market. Before I introduce these characters, though, I draw the history of the taxi market.

In that, I talk about taxi stands. Taxi stands work in the way of Thomas Schelling’s focal points. Passengers go there because they know empty taxis will go there. Taxi drivers looking for passengers go there because they know passengers looking for taxis will go there. This way, rather than waiting at a random place looking for either a passenger or a ride, going to the taxi stand is rational. And in that sense, taxi stands are a platforms.

In a way, railway platforms are platform in the same sense. Think of a train that wants to pick up passengers, and passengers who want to travel on a train. If there were no designated pick up points, trains would stop at random places, which passengers would have to guess. While engine drivers could see passengers waiting by the side, stopping at random places might have meant that the train would have had to go empty.

From this perspective, railway platforms act as platforms – they are focal points where trains and passengers come together. Passengers wait there because they know trains stop there, and vice versa. And helpfully, there is an actual physical platform that elevates passengers to the height of the train door so they can get on and off easily!

Isn’t this a wonderful way to have complicated a rather simple concept?