Sunday, March 24, 2013

Albert Hirschman and the Happiness of Pursuit

The following is the text of my remarks at a gathering in memory of Albert Hirschman, held earlier today at the Institute for Advanced Study. The event included moving recollections from members of his family, as well as tributes by Joan Scott, Jeremy Adelman, Michael Walzer, Amartya Sen, Annie Cot, Wolf Lepenies, William Sewell, James Wolfensohn, and Robbert Dijkgraaf. Fernando Henrique Cardoso could not attend but sent written remarks that were read out by Adelman. I'll update this post with links to the text or video of other speeches should any become available.

---

It’s an enormous privilege to have been invited to speak at this event in memory of Albert Hirschman. Unlike most of the other speakers here, I knew Albert only from a distance, based largely on his books and interviews. I met him in person just once, though I was fortunate enough to get to know Sarah a little during my year at the Institute.

Since my connection to Albert was largely through his writing, I’d like to speak about his love of language, his gift for expression, and his approach to the written word. To Albert, words were not merely vehicles for the transmission of ideas—they were objects to be played with and molded into structures in which one could perpetually take delight.

In 1993 Albert gave an interview to a group of Italian writers, which he later translated into English and published under the title Crossing Boundaries. I’d like to quote a segment of that interview that sums up very nicely both his playful relationship with language and the great originality of his ideas. This is what he said:
I enjoy playing with words, inventing new expressions. I believe there is much more wisdom in words than we normally assume.... Here is an example.  
One of my recent antagonists, Mancur Olson, uses the expression "logic of collective action" in order to demonstrate the illogic of collective action, that is, the virtual unlikelihood that collective action can ever happen. At some point I was thinking about the fundamental rights enumerated in the Declaration of Independence and that beautiful expression of American freedom as "the right to life, liberty, and the pursuit of happiness."I noted how, in addition to the pursuit of happiness, one might also underline the importance of the happiness of pursuit, which is precisely the felicity of taking part in collective action. I simply was happy when that play on words occurred to me.
This idea of the happiness of pursuit, the pleasure that one takes in collective action, was to be a central theme in his masterpiece Exit, Voice and Loyalty, published in 1970. Two centuries earlier, Adam Smith had spoken of our propensity to "truck, barter, and exchange one thing for another." Albert Hirschman spoke instead of a propensity to protest, complain, and generally "kick up a fuss." This articulation of discontent he called Voice.

Albert believed that voice was an important factor in arresting and reversing decline in firms, organizations, and states. Economists to that point had focused on a very different mechanism, namely desertion or exit, and had argued that greater competition, in the form of greater ease of exit, was a beneficial force in maintaining high levels of organizational performance.

Albert pointed out that there was a trade-off between exit and voice; that greater ease of exit could result in a stifling of voice as the individuals most inclined to protest and complain chose to depart instead. He also observed that loyalty, provided that it was not completely blind and uncritical, could serve to delay exit and thus create the space for voice to do its work.

What Albert did in Exit, Voice and Loyalty was nothing less than to reunite two disciplines, economics and political science, which had once been closely entwined but had drifted far apart over time. And he did this not by exporting the methods of economics to the analysis of politics, as others had done, but by emphasizing the importance of political activity within the economic sphere.

This kind of interdisciplinarity permeated all of Albert’s work. He described the idea of trespassing as "basic to his thinking." Crossing boundaries came naturally to him; he was too restless and playful to be confined to a single discipline. He was also an intellectual rebel, eager to question conventional wisdom whenever he found it wanting. In fact, he did so even when the conventional wisdom had been established by his own prior work. He referred to this as a propensity to self-subversion, which he called a "permanent trait of his intellectual personality."

I recall vividly and fondly my very first contact with Albert’s work. I had just begun graduate school, having never previously studied economics, and found myself in a course on the History of Economic Thought with the legendary Robert Heilbroner. It was Heilbroner’s book The Worldly Philosophers that had steered me to economics in the first place. And there on his syllabus, alongside Smith and Ricardo and Malthus, was Albert’s book The Passions and the Interests.

I recently went back and read this extraordinary book for a second time. The twentieth anniversary edition has a foreword by Amartya Sen, who considers it to be "among the finest" of Albert’s writings. Albert himself, in the preface to this edition, notes that it’s the one book that never fell victim to his propensity to self-subversion.

There’s a memorable passage in the book where Albert discusses Adam Smith’s claim that "order and good government" came to England as the unintended consequence of a growing taste for manufactured luxuries among the feudal elite. They "bartered their whole power and authority," says Smith, for the "gratification of… vanities… for trinkets and baubles, better fit to be the playthings of children than the serious pursuits of man." Having squandered their wealth in this manner, they could no longer support their vast armies of retainers, and became incapable of "disturbing the peace" or "interrupting the regular execution of justice."

But Albert was skeptical that the feudal lords had been quite so blind to their long-term interests. He felt that Smith, always eager to uncover the unintended effects of human action, had overreached this time. And he expressed this thought as follows:
One cannot help feeling that in this particular instance, Smith overplayed his Invisible Hand.
I can just imagine the smile that spread across Albert’s face when he came up with that turn of phrase.

Albert’s work was expansive and visionary, bold and audacious, breathtakingly original and creative. But most of all, it was playful and gently irreverent. He demonstrated to us, by his own example, the happiness of intellectual pursuit. For that, more than anything else, I’ll always be grateful.

Monday, March 11, 2013

A Prediction Market Mystery

The peer-to-peer prediction market Intrade ceased operations yesterday and closed out all open positions without notice. Visitors to the site were greeted with the following mysterious message (emphasis added):
With sincere regret we must inform you that due to circumstances recently discovered we must immediately cease trading activity on www.intrade.com. 
These circumstances require immediate further investigation, and may include financial irregularities which in accordance with Irish law oblige the directors to take the following actions:
  • Cease exchange trading on the website immediately.
  • Settle all open positions and calculate the settled account value of all Member accounts immediately.
  • Cease all banking transactions for all existing Company accounts immediately.
During the upcoming weeks, we will investigate these circumstances further and determine the necessary course of action. 
To mitigate any further risk to members’ accounts, we have closed and settled all open contracts at fair market value as of the close of business on March 10, 2013, in accordance with the Terms and Conditions of our customers’ use of the website. You may view your account details and settled account balances by logging into the website. 
At this time and until further notice, it is not possible to make any payments to members in accordance with their settled account balance until the investigations have concluded.
Translation: all open contracts have been closed out at current prices, account balances now reflect only cash positions, and no withdrawals can be made until further notice. Not a penny will be paid out to any member for the time being, no matter how large their cash balance may be.

What on earth is going on? My best guess is that the margin posted by traders was not held, as it should be, in segregated accounts separate from company funds. When bets are made on this market, both parties must post margin equal to their worst-case loss, so that neither is subject to counterparty risk. In effect, each party is taking a position against the exchange, but these positions are exactly offsetting so the exchange bears no risk. To ensure that all promised payments can be made, these funds must be held in the form of cash, insured deposits, or safe dollar-denominated securities such as Treasury bills. They cannot be invested in risky assets, and cannot be used for the payment of salaries or expenses.

All this was made clear in the exchange's so-called Trust and Security Statement:
Segregated Funds: Your funds are held in segregated accounts with banks in Ireland, and are segregated from Intrade's own corporate funds. 
Safer by Design: If the Dow Jones crashes, the New York Stock Exchange doesn't go bankrupt. In the same way, intrade doesn't lose money when an unusual result arises. Whenever you trade, intrade will 'freeze' sufficient money in your account to cover your potential losses. If you lose, we simply transfer the already frozen money from your account to a winning customer account. If you win, we pay your winnings from a losing customer account.
While this design is safe in theory, there was no mechanism in place to ensure that these commitments would, in fact, be met. When Intrade closed its doors to US residents in November, it did so in response to an action by the CFTC. I wondered at the time whether there was regulatory concern about the segregation of funds:
Even though the exchange claims to keep this margin in segregated accounts, separate from company funds, there is always the possibility that its deposits are not fully insured and could be lost if the Irish banking system were to collapse. These losses would ultimately be incurred by traders, who would then have very limited legal recourse.
Similar concerns were raised in an exchange with Dave Pennock on twitter. I thought at the time that the biggest risk came from failures in the Irish banking system, and discounted the possibility that trader margin could be deliberately co-mingled with company funds, invested in risky securities, or simply embezzled. This may have been too optimistic a view, especially given the precedent of MF Global.

If some funds have been diverted or lost, then traders face the prospect of receiving less than par on their cash balances when withdrawals eventually resume. And even if they do not suffer eventual losses, the fact that their funds are frozen for an extended period itself imposes an opportunity cost. If there's a lesson in all this, it is that markets cannot exist without trust, and trust cannot be sustained indefinitely without some sort of oversight and regulation. Reputational effects alone are simply not enough.

Friday, March 01, 2013

Why Do Groupon Campaigns Damage Yelp Ratings?

One of the many benefits of visiting Microsoft Research this semester is that I get to attend some interesting talks by computer scientists working with social and economic data. One in particular this week turned out to be extremely topical. The paper was on "The Groupon Effect on Yelp Ratings" and it was presented by Giorgios Servas Zervas.

The starting point of the analysis was this: the Yelp ratings of businesses who launch Groupon campaigns suffer a sharp and immediate decline which recovers only gradually over time, with peristent effects lasting for well over a year. The following chart sums it up:


The trend line is a 30 day moving average, but re-initialized on the launch date (so the first few points after this date average just a few observations). There is a second sharp decline after about 180 days, as the coupons are about to expire. The chart also shows the volume of ratings, which surges after the launch date. Part of the surge is driven by raters who explicitly reference Groupon (the darker volume bars). But not all Groupon users identify as such in their reviews, and about half the increase in ratings volume comes from ratings that do not reference Groupon.

As is typical of computer scientists working with social data, the number of total observations is enormous. Almost 17,000 daily deals from over 5,000 businesses in 20 cities over a six month period are included, along with review histories for these businesses both during and prior to the observational window. In addition, the entire review histories of those who rated any of these businesses during the observational window were collected, expanding the set of reviews to over 7 million, and covering almost a million distinct businesses in all.

So what accounts for the damage inflicted on Yelp ratings by Groupon campaigns? The authors explore several hypotheses. Groupon users could be especially harsh reviewers regardless of whether or not they are rating a Groupon business. Businesses may be overwhelmed by the rise in demand, resulting in a decline in quality for all customers. The service provided to Groupon users may be worse than that provided to customers paying full price. Customer preferences may be poorly matched to businesses they frequent using Groupons. Or the ratings prior to the campaign may be artificially inflated by fake positive reviews, which get swamped by more authentic reviews after the campaign. All of these seem plausible and consistent with anecdotal evidence.

One hypothesis that is rejected quite decisively by the data is that Groupon users tend to be harsh reviewers in general. To address this, the authors looked at the review histories of those who identified Groupon use for the businesses in the observational window. Most of these prior reviews do not involve Groupon use, which allows for a direct test of the hypothesis that these raters were harsh in general. It turns out that they were not. Groupon users tend to write detailed and informative reviews that are more likely to be considered valuable, cool and funny by their peers. But they do not rate businesses without Groupon campaigns more harshly than other reviewers.

What about the hypothesis of businesses being overwhelmed by the rise in demand? Since only about half the surge in reviews comes from those who explicitly reference Groupon, the remaining ratings pool together non-Groupon customers with those who don't reveal Groupon use. This makes a decline in ratings by the latter group hard to interpret. John Langford (who was in the audience) noted that if the entire surge in reviews could be attributed to Groupon users, and if undeclared and declared users had the same ratings on average, then one could infer the effect of the campaign on the ratings of regular customers. This seems worth pursuing.

Anecdotal evidence on discriminatory treatment of customers paying discounted prices is plentiful (the authors mention the notorious FTD flowers case for instance). If mistreatment of coupon-carrying customers by a few bad apples were bringing down the ratings average, then a campaign should result in a more negatively skewed distribution of ratings relative to the pre-launch baseline. The authors look for this shift in skewness and find some evidence for it, but the effect is not large enough to account for the entire drop in the average rating.

To test the hypothesis that ratings prior to a campaign are artificially inflated by fake or purchased reviews, the authors look at the rate at which reviews by self-identified Groupon users are filtered, compared with the corresponding rate for reviews that make no mention of Groupon. (Yelp allows filtered reviews to be seen, though they are harder to access and are not used in the computation of ratings). Reviews referencing Groupon are filtered much less often, suggesting that they are more likely to be authentic. If Yelp's filtering algorithm is lenient enough to let a number of fake reviews through, then the post-campaign ratings will be not just more numerous but also more authentic and less glowing.

Finally, consider the possibility of a mismatch between the preferences of Groupon users and the businesses whose offers they accept. To look for evidence of this, the authors consider the extent to which reviews associated with Groupon use reveal experimentation on the part of the consumer. This is done by comparing the business category and location to the categories and locations in the reviewer's history. Experimentation is said to occur when the business category or zipcode differs from any in the reviewer's history. The data provide strong support for the hypothesis that individuals are much more likely to be experimenting in this sense when using a Groupon than when not. And such experimentation could plausibly lead to a greater incidence of disappointment.

This point deserves further elaboration. Even without experimentation on categories or locations, an individual who accepts a daily deal has a lower anticipated valuation for the product or service than someone who pays full price. Even if the expectations of both types of buyers are met, and each feels that they have gotten what they paid for, there will be differences in the ratings they assign. To take an extreme case, if the product were available for free, many buyers would emerge who consider the product essentially worthless, and would rate it accordingly even if their expectations are met.

There may be a lesson here for companies contemplating Groupon campaigns. Perhaps the Yelp rating would suffer less damage if the discount itself were not as steep. At present there is very little variation in discounts, which are mostly clustered around 50%. So there's no way to check whether smaller discounts actually result in better ratings relative to larger discounts. But it certainly seems worth exploring, at least for businesses that depend on strong ratings to thrive.

The Groupon strategy of prioritizing growth above earnings had been criticized on the grounds that there are few barriers to entry in this industry, and no network externalities that can protect an incumbent from competition. But if the link between campaigns and ratings can't be broken, there may be deeper problems with the business model than a change of leadership or strategy can solve.